Chemotherapy versus best supportive care for extensive small cell lung cancer

  • Review
  • Intervention

Authors


Abstract

Background

Combination chemotherapy has been the mainstay of treatment for extensive stage small celI lung cancer (SCLC) over the last 30 years, even though it only gives a short prolongation in median survival time. The main goal for these patients should be palliation with the aim of improving their quality of life.

Objectives

To determine the effectiveness of first-line chemotherapy versus placebo or best supportive care (BSC) in prolonging survival in patients with extensive SCLC at diagnosis and the effectiveness of second-line chemotherapy at relapse or progression after first-line chemotherapy compared with BSC or placebo in prolonging survival in patients with extensive SCLC; as well as to evaluate the adverse events of treatment and the quality of life of patients.

Search methods

This is the second update of the review. MEDLINE (1966 to October 2013), EMBASE (1974 to October 2013), and the Cochrane Central Register of Controlled Trials (CENTRAL) (2012, Issue 3) were searched. Experts in the field were contacted.

Selection criteria

Phase III randomised controlled trials in which any chemotherapy treatment was compared with  placebo or BSC in patients with extensive SCLC, as first-line or second-line therapy at relapse.

Data collection and analysis

Two authors independently extracted data and assessed study quality. We resolved disagreements by discussion. Additional information was obtained from one study author.

Main results

Two studies of unclear risk of bias were included for first-line chemotherapy. A total of 88 men under 70 years with good performance status were randomised to receive either supportive care, placebo infusion or ifosfamide. Ifosfamide gave an extra mean survival of 78.5 days compared with supportive care or placebo infusion. Partial tumour response was greater with the active treatment. Toxicity was only seen in the chemotherapy group and quality of life was only assessed at the beginning of treatment. The quality of the evidence for overall survival and adverse effects was very low.

Three studies of moderate risk of bias were included for second-line chemotherapy at relapse (one identified in the last search). A total of 932 men and women under 75 years and any performance status were randomised to receive either methotrexate-doxorubicin, topotecan, or picoplatin versus symptomatic treatment or BSC. The methotrexate-doxorubicin treatment gave a median survival of 63 days longer than in the symptomatic-treatment group for patients allocated to receive four cycles of first-line chemotherapy, and 21 days longer for patients allocated to receive eight cycles of first-line chemotherapy.

Treatment with topotecan gave a median survival of 84 days longer than in the BSC group (log-rank P = 0.01). The adjusted hazard ratio (HR) for overall survival was 0.61 (95% CI 0.43 to 0.87). Treatment with picoplatin gave a median survival time of six days longer than BSC (HR 0.817, 95% CI 0.65 to 1.03, P = 0.0895). A meta-analysis of topotecan and picoplatin gave a HR of 0.73 (95% CI 0.55 to 0.96, P = 0.03; low-quality evidence).

Partial or complete response in the methotrexate-doxorubicin group was 22.3%. Five patients (7%, 95% CI 2.33 to 15.67) showed a partial response with topotecan. No data were provided about tumour response in the picoplatin study. Toxicity was worst in the chemotherapy group (moderate-quality evidence). Quality of life was better in the topotecan group and was not measured in the methotrexate-doxorubicin and picoplatin studies (low-quality evidence).

Authors' conclusions

Two small RCTs from the 1970s suggest that first-line chemotherapeutic treatment (based on ifosfamide) may provide a small survival benefit (less than three months) in comparison with supportive care or placebo infusion in patients with advanced SCLC. However platinum-based combination chemotherapy regimens have been shown to increase complete response rates when compared to non-platinum chemotherapy regimens with no significant difference in survival, and so these are currently the standard first-line treatment for patients with SCLC.

Second-line chemotherapy at relapse or progression may prolong survival for some weeks in relation to BSC. Nevertheless, the impact of first-line chemotherapy on quality of life, older patients, women and patients with poor prognosis is unknown and the benefits of second-line chemotherapy are also unclear for older people. Globally, the evidence on which these conclusions are based is very scarce and of uncertain or low quality, which calls for well-designed, controlled trials to further evaluate the trade-offs between benefits and risks of different chemotherapeutic schedules in patients with advanced SCLC.

Résumé scientifique

Chimiothérapie ou soins palliatifs pour le cancer bronchique à petite cellules disséminé

Contexte

Depuis 30 ans, le cancer bronchique à petites cellules (CBPC) disséminé est principalement traité par polychimiothérapie, bien que ce traitement ne permette pas d'obtenir qu'une prolongation réduite de la moyenne de survie. L'objectif principal pour ces patients pourrait être palliatif afin d'améliorer leur qualité de vie.

Objectifs

Déterminer l'efficacité de la chimiothérapie de première ligne versus un placebo ou aux soins palliatifs (SP) pour prolonger la survie chez les patients atteints de CBPC disséminé au diagnostic et de la chimiothérapie en deuxième intention après une rechute ou de progression de la maladie après la chimiothérapie de première ligne par rapport aux SP ou au placebo ainsi qu'évaluer les événements indésirables du traitement et la qualité de vie des patients.

Stratégie de recherche documentaire

Ceci est la deuxième mise à jour de la revue. Ils ont été consultés MEDLINE (de 1966 à octobre 2013), EMBASE (de 1974 à octobre 2013), et le registre Cochrane des essais contrôlés (CENTRAL) (2012, numéro 3). Nous avons pris contact avec des experts.

Critères de sélection

Essais contrôlés randomisés en phase III dans lesquels un traitement avec chimiothérapie était comparé à un placebo ou aux SP chez des patients atteints de CBPC disséminé, en tant que traitement de première intention ou comme traitement de deuxième intention après une rechute.

Recueil et analyse des données

Deux auteurs ont extrait les données indépendamment et évalué la qualité des études. Les désaccords ont été résolus par discussion. Des informations supplémentaires ont été obtenues auprès de l'auteur d'une étude.

Résultats principaux

Deux études avec risque incertain de biais ont été inclus pour la chimiothérapie de première ligne. Un total de 88 hommes âgés de moins de 70 ans avec en bon état de performances ont été randomisés pour recevoir soit des soins palliatifs soit une perfusion de placebo ou d'ifosfamide. L'ifosfamide, a donné un supplément de survie moyen de 78,5 jours par rapport aux soins palliatifs ou à la perfusion d'un placebo. La réponse partielle tumorale a été supérieure avec le traitement actif. La toxicité n'a été observée que dans le groupe avec chimiothérapie et la qualité de vie a été évaluée seulement au début du traitement. La qualité des preuves pour la survie globale et les effets indésirables était très faible.

Trois études avec risque modéré de biais ont été inclus avec une chimiothérapie en deuxième intention après une rechute (une identifiée dans la dernière recherche). Un total de 932 hommes et femmes de moins de 75 ans et sans aucun indice de performance étaient randomisés pour recevoir soit Méthotrexate-Doxorubicine ou Topotécan soit Picoplatin contre un traitement symptomatique ou SP. Le traitement par Méthotrexate-Doxorubicine a permis une prolongation de la moyenne de survie 63 jours par rapport au groupe avec un traitement symptomatique pour les patients recevant quatre cycles de chimiothérapie de première ligne et 21 jours pour les patients recevant huit cycles de chimiothérapie de première ligne.

Le traitement avec Opotécan a permis de prolonger la survie moyenne 84 jours par rapport au groupe SP (Test du logrank, P =0,01). Le taux de hazard ajusté (TH) pour la survie globale était de 0,61 (IC à 95%, de 0,43 à 0,87). Le traitement avec Picoplatin donnait un temps de survie moyen de six jours par rapport aux SP (HR 0,817, IC à 95%,de 0,65 à 1,03; P =0,0895). Une méta-analyse de Topotécan et Picoplatin produisait un TH de 0,73 (IC à 95%, de 0,55 à 0,96, P =0,03; preuves de faible qualité).

La réponse partielle ou complète dans le groupe Méthotrexate-Doxorubicine était de 22,3%. Avec le Topotécan, une réponse partielle a été rapportée chez cinq patients (7%, IC à 95%, entre 2,33 et 15,67). Aucune donnée n'a été fournie sur la réponse tumorale dans l"étude avec Picoplatin. La toxicité était plus importante dans le groupe avec chimiothérapie (preuves de qualité moyenne). La qualité de vie était meilleure dans le groupe Topotécan et n'était pas évaluée dans les études avec Méthotrexate-Doxorubicine ni avec Picoplatin (preuves de faible qualité).

Conclusions des auteurs

Deux ECR de petite taille dans les années 1970 suggèrent que la chimiothérapie en première ligne de traitement (sur la base d'Ifosfamide) pourrait apporter un petit bénéfice en survie (moins de trois mois) en comparaison avec les soins palliatifs ou la perfusion de placebo chez des patients atteints de CBPC disséminé. Cependant la polychimiothérapie à base de platines a été démontré augmenter les taux de réponse complète par rapport à des régimes de chimiothérapie sans platines avec absence de différence significative dans la survie et ce sont ainsi, à présent, le traitement standard de première intention pour les patients atteints de CBPC.

La chimiothérapie en deuxième intention après une rechute ou de progression de la maladie peut prolonger la survie quelques semaines par rapport aux SP. Néanmoins, l'impact de la chimiothérapie de première ligne sur la qualité de vie, chez les patients âgés, les femmes et les patients présentant un mauvais pronostic est inconnu et les bénéfices de la chimiothérapie en deuxième intention sont aussi peut évidents pour les personnes âgées. En général, des preuves sur lesquelles ces conclusions sont basées sont très rares et incertaines ou de faible qualité, ce qui fait appel à des essais contrôlés bien planifiés afin de mieux évaluer le rapport bénéfices/risques des différents schémas chimiothérapeutiques chez les patients atteints de CBPC disséminé.

Resumo

Quimioterapia versus melhores cuidados de suporte para câncer de pulmão de pequenas células avançado

Introdução

A quimioterapia combinada tem sido a base do tratamento para o câncer de pulmão de pequenas células (CPPC) avançado pelos últimos 30 anos, mesmo que esta apenas possibilite um pequeno prolongamento do tempo médio de sobrevida. O principal objetivo para estes pacientes deve ser paliativo visando um ganho na qualidade de vida

Objetivos

Determinar a efetividade da quimioterapia de primeira linha versus placebo ou melhores cuidados de suporte (MCS) em prolongar a sobrevida em pacientes com CPPC avançado na época do diagnóstico e a efetividade da quimioterapia de segunda linha na recidiva ou progressão após quimioterapia de primeira linha comparada com MCS ou placebo em prolongar a sobrevida de pacientes com CPPC avançado; bem como avaliar os efeitos adversos do tratamento e a qualidade de vida dos pacientes.

Métodos de busca

Esta é a segunda atualização da revisão. MEDLINE (1966 a Outubro de 2013), EMBASE (1974 a Outubro de 2013), e o Registro Central Cochrane de Ensaios Controlados (CENTRAL) (2012, Edição 3) foram pesquisados. Autoridades no assunto foram contatadas.

Critério de seleção

Ensaios clínicos randomizados e controlados fase III em que qualquer tratamento quimioterápico foi comparado com placebo ou MCS em pacientes com CPPC avançado, como terapia de primeira linha ou segunda linha na recidiva.

Coleta dos dados e análises

Dois autores independentemente extraíram as informações e avaliaram a qualidade dos estudos. Nós resolvemos as discordâncias através de discussão. Informação adicional foi obtida junto a um autor de estudo.

Principais resultados

Dois estudos nos quais o risco de viés não foi evidente foram incluídos para quimioterapia de primeira linha. Um total de 88 homens com menos de 70 anos, com bom estado geral foram randomizados para receber ou cuidados de suporte, infusão de placebo ou ifosfamida. Ifosfamida deu uma sobrevida média adicional de 78,5 dias, em comparação com cuidados de suporte ou infusão placebo. Resposta parcial do tumor foi maior com o tratamento ativo. Toxicidade só foi observada no grupo de quimioterapia e qualidade de vida foi avaliada apenas no início do tratamento. A qualidade da evidência para a sobrevida geral e efeitos adversos foi muito baixa.

Três estudos com risco moderado de viés foram incluídos para quimioterapia de segunda linha na recidiva (um identificado na última busca). Um total de 932 homens e mulheres com menos de 75 anos e de qualquer nível de estado geral foram randomizados para receber metotrexato-doxorrubicina, topotecano, ou picoplatina versus tratamento sintomático ou MCS. O tratamento com metotrexato-doxorrubicina deu uma sobrevida média de 63 dias a mais do que no grupo de tratamento sintomático para pacientes alocados para receber quatro ciclos de quimioterapia de primeira linha, e 21 dias a mais para os pacientes alocados para receber oito ciclos de quimioterapia de primeira linha.

O tratamento com topotecano deu uma sobrevida média de 84 dias a mais do que no grupo MCS (log-rank P = 0,01). A razão de risco ajustada (RR) para a sobrevida geral foi de 0,61(IC 95% 0,43-0,87). O tratamento com picoplatina deu um tempo médio de sobrevida de seis dias a mais do que MCS (HR 0,817, 95% CI 0,65-1,03, P = 0,0895). Uma meta-análise de topotecano e picoplatina deu uma RR de 0,73 (95% CI 0,55-0,96, P = 0,03; evidência de baixa qualidade).

Resposta parcial ou completa no grupo metotrexato-doxorrubicina foi de 22,3%. Cinco pacientes (7%, IC 95% 2,33-15,67) apresentaram uma resposta parcial com topotecano. Não foram fornecidos dados sobre a resposta do tumor no estudo picoplatina. A toxicidade foi pior no grupo de quimioterapia (qualidade de evidência moderada). A qualidade de vida foi melhor no grupo de topotecano e não foi medida nos estudos com metotrexato-doxorrubicina e picoplatina (qualidade de evidência baixa).

Conclusão dos autores

Dois pequenos ensaios clínicos da década de 1970 sugerem que o tratamento quimioterápico de primeira linha (com base na ifosfamide) pode fornecer um pequeno benefício na sobrevida (menos de três meses) em comparação com cuidados de suporte ou infusão de placebo em doentes com CPPC avançado.comparação com cuidados de suporte ou infusão de placebo em doentes com CPPC avançado. No entanto protocolos de quimioterapia de combinação à base de platina têm mostrado um aumento nas taxas de resposta completa quando comparadas a protocolos de quimioterapia sem platina, sem diferença significativa na sobrevida, e é por isso que são atualmente o tratamento padrão de primeira linha para pacientes com CPPC.

Quimioterapia de segunda linha na recidiva ou progressão pode prolongar a sobrevida em algumas semanas em relação aos MCS. No entanto, o impacto da quimioterapia de primeira linha na qualidade de vida, em pacientes idosos, mulheres e pacientes com mau prognóstico é desconhecido e os benefícios da quimioterapia de segunda linha também não são claros para os idosos. De modo geral, as evidências em que se baseiam essas conclusões são muito escassas e de qualidade duvidosa ou baixa, o que mostra a necessidade de ensaios clínicos bem desenhados, para avaliar melhor a comparação entre os benefícios e os riscos dos diferentes esquemas de quimioterapia em doentes com CPPC avançado.

Notas de tradução

Traduzido por: Cláudio José Rubira, Unidade de Medicina Baseada em Evidências da Unesp, Brazil Contato:portuguese.ebm.unit@gmail.com

Plain language summary

Chemotherapy (anticancer drugs) for patients with advanced small cell lung cancer

SmalI cell lung cancer accounts for nearly a quarter of all new cases of lung cancer. This cancer is often diagnosed in an advanced stage, which means that it has spread to the brain, liver, bone or bone marrow, and most patients die in the first year after diagnosis. This review found that first-line chemotherapy (anticancer drugs) may prolong the survival of patients with advanced small cell lung cancer for some months when compared to supportive care, although the effect of this treatment on quality of life is unknown. The benefit of a new treatment (second-line chemotherapy) when the disease has progressed or relapsed was even smaller, and the potential survival gain of some weeks must be balanced against its possible secondary effects. Since the available studies were scarce and of variable quality, more clinical trials are needed to assess and better inform patients about the real effectiveness of chemotherapy in advanced small cell lung cancer.

Résumé simplifié

La chimiothérapie (médicaments anticancéreux) pour les patients atteints de cancer bronchique à petites cellules disséminé

Le cancer bronchique à petites cellules représente près d'un quart de tous les nouveaux cas de cancer du poumon. Ce cancer est souvent diagnostiqué dans un stade avancé, ce qui signifie qu' il s' est propagé dans le cerveau, le foie, les os ou de la moelle osseuse et la plupart des patients décèdent dans la première année après le diagnostic. Cette revue a trouvé que la chimiothérapie en première ligne (médicaments anticancéreux) peut prolonger la survie des patients atteints de cancer bronchique à petites cellules avancé quleques mois par rapport aux soins palliatifs, bien que l'effet de ce traitement sur la qualité de vie est inconnu. Le bénéfice d'un nouveau traitement (chimiothérapie en deuxième intention) lorsque la maladie a progressé ou en rechute était encore plus faible et la possibilité de survie de quelques semaines doit être mise en balance avec les effets secondaires possibles. Puisque les études disponibles étaient rares et de qualité variable, des essais cliniques supplémentaires sont nécessaires pour évaluer et de mieux éclairer les patients concernant l'efficacité réelle de la chimiothérapie dans le cancer bronchique à petites cellules disséminé.

Notes de traduction

Traduit par: French Cochrane Centre 14th January, 2014
Traduction financée par: Ministère du Travail, de l'Emploi et de la Santé Français

Laički sažetak

Kemoterapija (antitumorski lijekovi) za pacijente s uznapredovalim rakom pluća malih stanica

Gotovo četvrtina novih slučajeva raka pluća je rak pluća malih stanica. Ovaj tip raka se često dijagnosticira u uznapredovaloj fazi, što znači da se proširio na mozak, jetru, kosti ili koštanu srž te većina pacijenata umre u prvoj godini nakon dijagnoze. Ovaj Cochrane sustavni pregled je utvrdio da prva linija kemoterapije (antitumorski lijekovi) može produljiti preživljenje pacijenata s uznapredovalim rakom pluća malih stanica za nekoliko mjeseci u usporedbi sa samom potpornom skrbi, iako je učinak liječenja kemoterapijom na kvalitetu života nepoznat. Prednost nove terapije (kemoterapija druge linije) kada je bolest napredovala ili se ponovo javila (relapsirala) je bila još manja, i potencijalno preživljenje od nekoliko tjedana više se mora razmotriti u svjetlu mogućih nuspojava. Budući je nađeno malo studija (2 studije o kemoterapiji kao prvoj liniji liječenja i 3 studije o kemoterapiji kao drugoj liniji liječenja nakon relapsa), koje su bile varijabilne kvalitete, potrebno je još kliničkih studija koje bi utvrdile i pružile više informacija pacijentima o stvarnoj učinkovitosti kemoterapije za rak pluća malih stanica.

Bilješke prijevoda

Hrvatski Cochrane
Preveo: Adam Galkovski
Ovaj sažetak preveden je u okviru volonterskog projekta prevođenja Cochrane sažetaka. Uključite se u projekt i pomozite nam u prevođenju brojnih preostalih Cochrane sažetaka koji su još uvijek dostupni samo na engleskom jeziku. Kontakt: cochrane_croatia@mefst.hr

Resumo para leigos

Quimioterapia (drogas anticancer) para pacientes com câncer de pulmão de pequenas células avançado

Câncer de pulmão de pequenas células é responsável por aproximadamente um quarto de todos os casos novos de câncer de pulmão Esse tipo de câncer é muitas vezes diagnosticado em um estágio avançado, o que significa que ele se espalhou para o cérebro, fígado, osso ou medula óssea , e a maioria dos pacientes morrem no primeiro ano após o diagnóstico Esta revisão constatou que a quimioterapia de primeira linha (drogas anticâncer) pode prolongar a sobrevida de pacientes com câncer de pulmão de pequenas células avançado por alguns meses quando comparados aos cuidados de suporte, embora o efeito deste tratamento na qualidade de vida é desconhecida O benefício de um novo tratamento (quimioterapia de segunda linha), quando a doença tenha avançado ou recidivado menos, e o ganho potencial de sobrevida de algumas semanas deve ser comparado com os seus possíveis efeitos secundários. Uma vez que os estudos disponíveis foram escassos e de qualidade variável, mais ensaios clínicos são necessários para avaliar e melhor informar aos pacientes sobre a real efetividade da quimioterapia em câncer de pulmão de pequenas células avançado.

Notas de tradução

Traduzido por: Cláudio José Rubira, Unidade de Medicina Baseada em Evidências da Unesp, Brazil Contato:portuguese.ebm.unit@gmail.com

Summary of findings(Explanation)

Summary of findings for the main comparison. First-line chemotherapy compared with BSC for extensive SCLC
  1. 1. The corresponding median survival time in extensive SCLC in the control and intervention groups has been obtained from Kokron 1977 and Kokron 1982

    2. Significant risk of bias across the studies

    3. Small sample sizes across the studies

First-line chemotherapy compared with BSC for extensive SCLC

Patient or population: patients with extensive SCLC without previous treatment

Settings: one hospital in Vienna

Intervention: first-line chemotherapy

Comparison: Supportive treatment

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
Supportive careFirst-line chemotherapy
Overall survivalThe mean survival ranged across control groups from
56-93 days 1
The mean survival in the intervention groups was
134-172 days 1
Not estimable65 (2 studies)⊕⊝⊝⊝
very low ,2,3
 
Adverse effects

Haematological: 0%

Leucopenia:0%

Vomiting and hair loss: 0%

Other: 0%

Haematological:   94.1%

Leucopenia:15% ifosfamide; 15.8% ifosfamide + CCNU

Vomiting and hair loss: 70% ifosfamide; 68.4% ifosfamide + CCNU

Other: 55% ifosfamide; 52.6% ifosfamide + CCNU

Not estimable65 (2 studies)⊕⊝⊝⊝
very low 2,3
 
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Summary of findings 2 Second-line chemotherapy at relapse or progression in extensive SCLC

Summary of findings 2. Second-line chemotherapy at relapse or progression in extensive SCLC
  1. 1. The corresponding median survival time in extensive SCLC in the control group has been obtained from O'Brien 2006 and Ciuleanu 2010

    2. The corresponding median survival time in extensive SCLC in the intervention group has been obtained from O'Brien 2006 and Ciuleanu 2010

    3. The difference in rate of deterioration was obtained through the EQ-5D (EuroQol-5-Dimensions Health) questionnaire, consisting of five health status dimensions: mobility, self-care, usual activities, pain and discomfort, anxiety and depression, with rating of 1 (no problem) to 3 (extreme problem). Patients were evaluated before each course of second-line chemotherapy, or at each visit in the BSC arm.

    The PSA (Patient Self Assessment) questionnaire rating scale for severity is from 1 (not at all), to 4 (very much)

    4. Some significant risk of bias across the studies

    5. Some significant risk of other bias in one study

Second-line chemotherapy compared with BSC for extensive SCLC

Patient or population: patients with extensive SCLC who relapsed or progressed after first-line chemotherapy

Settings: multi-institutional-multicentre

Intervention: second-line chemotherapy

Comparison: BSC

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
BSCSecond-line chemotherapy
Overall survivalThe median survival ranged across control groups from 97.3-138 days1The median survival in the intervention groups was 144-181.3 days2

HR: 0.73

(0.55, 0.96)

542

(2 studies)

⊕⊕⊝⊝
low 4,5
 
Toxic death0%

6% topotecan

0% picoplatin

Not estimable

542

(2 studies)

⊕⊕⊕⊝
moderate 4
 
Adverse effects

Dyspnoea 3%

Fatigue 4%

Non sepsis infection 12%

sepsis 1%

Haematological: 18%-61% neutropenia, 38%-41% Thrombocytopenia, 25%-29% anaemia,

Non sepsis infection 14%

Sepsis 4%

Asthenia 11%

Not estimable

542

(2 studies)

⊕⊕⊕⊝
moderate 4
 
Quality of lifeSee footnotes 3See footnotes 3

Difference in rate of deterioration per 3-month intervals in the EQ-5D score: 0.15 (95% CI 0.05, 0.25)

In the PSA questionnaire:

shortness of breath RR: 2.18 (1.09 to 4.38)

interference with sleep RR: 2.16 (1.15 to 4.06)

fatigue RR: 2.29 (1.25 to 4.19)3

141

(1 study)

⊕⊕⊝⊝

low 4,5

 
*The basis for the assumed risk is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
BSC: best supportive care; CI: Confidence interval; RR: Risk ratio; HR: hazard ratio
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Background

Description of the condition

Small cell Iung cancer (SCLC) accounts for 15% to 17% of new cases of lung cancer among males and females respectively (Riaz 2012), with a gradual decline in the proportion of male patients compared to female patients in the last two decades. Women comprised 28% of all cases in 1973 yet by 2002 this proportion had risen to 50% (Weatley-Price 2010). As a result of increasing life expectancy, about 32% of newly diagnosed SCLC cases occur in patients older than 70 years of age and 10% of cases are diagnosed in patients older than 80 years of age (Owonikoko 2007). Early diagnosis of SCLC is uncommon. Between 60% and 70% of patients have extensive disease at diagnosis with metastases involving one or more sites, such as the brain, liver, bone or bone marrow (Carney 2002). Untreated SCLC is lethal within two to three months of diagnosis in most patients, and survival beyond one year is rare (De Vita 2000; Vrdoljak 2001; Zelen 1973). The majority of patients with extensive disease will relapse after treatment, usually within the first year after diagnosis. The two-year cumulative recurrence rate is 75% in patients with limited disease and nearly 100% in patients with extensive disease. The lungs are the most frequent site of initial recurrence, followed by the liver, bone, and the brain (Gralla 2004; Sugiyama 2008). Local treatments such as resection and radiotherapy have a limited effect on advanced disease (Souhami 1990) so the most widely accepted option for treating advanced SCLC is chemotherapy. Current evidence does not support screening for lung cancer with any modality, including chest radiography, sputum cytology, or spiral computed tomography (Manser 2010). There is no evidence either for recommending vitamin supplements for the prevention of lung cancer (Cortés-Jofré 2012).

The effect of chemotherapy with single agents was evaluated in controlled studies by the Veteran Administration Lung Group, which showed the effectiveness of cyclophosphamide in the treatment of SCLC when compared with placebo (Green 1969). Following on from these studies, combination chemotherapy became the mainstay of treatment for patients with extensive-stage SCLC in the 1970s (Ihde 1992). In the 1980s, the most widely used combination of drugs for initial treatment of extensive-stage SCLC was cyclophosphamide, doxorubicin, and vincristine, which produced partial or complete tumour remissions in 50% to 85% of patients and increased the median survival time by between nine and 12 months in unselected patient groups (De Wet 1994; Hollen 1994a; Hollen 1994b; Oze 2009; Souhami 1990). In the late 1980s, a combination regime of cisplatin and etoposide was introduced. Despite diverse strategies for the treatment of these patients, the results of phase III trials over the past 30 years have shown only a two month prolongation in median survival time for patients treated with different regimens (Chutte 1999), or a 0.021 month (0.63 day) increase in median survival time per year (Oze 2009), with 5% to 10% surviving two years and only 1% of patients achieving long-term disease-free survival (Govindan 2006; Hanna 2002; Lassen 1998; Wolf 2004). Overall, women have had a significantly longer median survival than men (Weatley-Price 2010). The lowest survival rates are in the very elderly, 80 years or older (Owonikoko 2007).

A large number of clinical trials comparing different regimens, schedules, and dose intensities of agents have been conducted in advanced SCLC. Comparator regimens have varied greatly between trials, not only with respect to agents but also in dosage, scheduling, and patient and disease factors. The assimilation of findings from this large range of heterogeneous trials is therefore difficult. To date no consistent findings have emerged as to the optimal agent combination or as to the effectiveness of increased dose intensity (Carney 2002). Recommended treatment regimens have varied with time and at present combinations of platinum are most often indicated for advanced disease (Seidenfeld 2006). A recent systematic review of platinum versus non-platinum chemotherapy regimens for SCLC did not find a statistically significant benefit in survival or overall tumour response for platinum-based therapy (Amarasena 2009). However, platinum-based chemotherapy regimens did increase complete response rates, at the cost of higher adverse events. Patients treated with chemotherapy can be categorized as having refractory disease where the disease progresses through first-line treatment; resistance disease, which includes those patients with an initial response to treatment but the disease then progresses within three months of completing treatment; and sensitive disease where patients who have a relapse-free interval of at least three months from completion of treatment receive re-treatment with first-line therapy as the standard approach (Garassino 2011). Second-line chemotherapy offered to patients at relapse may produce tumour regression but the evidence for a clinical benefit is limited. The degree of benefit seems to be related to the type of first-line chemotherapy used, the length of the treatment-free interval, the extent of disease at relapse, and the performance status of the individual (Sundstr¢m 2005).Thus, until better treatments are developed, the main goal of the therapy for extensive-stage SCLC should be adequate palliation and improvement of overall quality of life.

Our purpose was to review the evidence for the effectiveness of first-line and second-line chemotherapy at relapse or progression in extensive disease SCLC. This is the second update of the review.

Description of the intervention

The purpose of this review was to evaluate the effectiveness of first-line chemotherapy and second-line chemotherapy at relapse or progression after first-line chemotherapy in patients with extensive SCLC compared with best supportive care (BSC) or placebo treatment.

How the intervention might work

Benefits of first-line or second-line chemotherapy in advanced SCLC patients should include a significantly longer survival than for those treated with best supportive care (BSC) or placebo. The significant improvement in survival should be accompanied by manageable toxicity and an improvement or stabilisation in the quality of life.

Why it is important to do this review

This review of first-line chemotherapy compared with BSC or placebo treatment and second-line chemotherapy at relapse or progression after first-line chemotherapy is the second update of a previous review conducted in 2003 (Agra 2003) and updated in 2009 (Pelayo 2009). After completion of the original review several questions were raised about future research into the benefits of chemotherapy in extensive SCLC. It was suggested that the feasibility and ethics of undertaking studies comparing active treatment with BSC in patients with poor prognosis should be evaluated. Consideration of the risks and benefits of first and second-line chemotherapy in this subgroup of patients would help to define clearer and more explicit criteria about when to stop chemotherapy or when not to administer second-line treatment with the objective of preserving quality of life. Therefore, the focus of this review is to evaluate chemotherapy compared with BSC in patients with poor prognosis that present with extensive disease at diagnosis or have experienced relapse or progression after first-line treatment. All are patients with an extremely low short-term survival rate.

Objectives

1. To determine the effectiveness of first-line chemotherapy versus placebo or best supportive care (BSC) in prolonging survival in patients with extensive SCLC at diagnosis.

2. To determine the effectiveness of second-line chemotherapy at relapse or progression after first-line chemotherapy compared with BSC or placebo in prolonging survival in patients with extensive SCLC.

3. To evaluate the adverse events of treatment as well as the quality of life of patients with extensive SCLC who were treated with first or second-line chemotherapy.

Methods

Criteria for considering studies for this review

Types of studies

Phase III randomised controlled trials in patients with extensive stage SCLC. Studies were eligible for inclusion if they were described by the authors as ‘randomised’ anywhere in the manuscript.
There was no language restriction in considering studies eligible for the review.

We excluded phase I and phase II randomised controlled trials.

Types of participants

Patients of any age with histologically proven SCLC carcinoma staged as extensive or metastatic disease after completion of clinical and diagnostic staging studies. Extensive or metastatic SCLC cancer means that the tumour has spread beyond one hemithorax, including the ipsilateral mediastinal and supraclavicular fossa lymph nodes. All patients with distant metastases were considered to have extensive-stage disease.

For studies of first-line chemotherapy, patients that were included were those that had not received previous treatment (chemotherapy, radiotherapy, hormonal treatment, growth factor, immunotherapy, surgery). For studies of second-line chemotherapy, patients were included if they had received previous chemotherapy treatment and after relapse or progression they received a different chemotherapy regimen. Patients who had received previous radiotherapy, surgery, hormonal treatment, growth factor, or immunotherapy were excluded.

Types of interventions

Studies in which any chemotherapy regimen (any single agent or combination of agents, at any dosage, for any duration or number of cycles) was compared with a control group who received placebo or BSC, and in which there was no concomitant or sequential radiotherapy with curative intention. BSC measures included pain medications and other non-specific palliative treatments (including radiotherapy with explicitly palliative intention, <5 0 Gy) that were intended to palliate a particular symptom.

We excluded studies in which any anticancer treatment other than chemotherapy (radiotherapy or chemoradiotherapy, surgery, hormonal treatment, growth factors, immunotherapy) was compared with either BSC or a placebo intervention.

Types of outcome measures

Primary outcomes

I) Overall survival, measured as the hazard ratio or median or mean values with confidence intervals

Secondary outcomes

II) Survival at 6,12, 24 months

III) Complete* and partial** response (see below)

IV) Toxicity, toxic deaths and any other measures of toxicity found to be recorded consistently across studies

V) Quality of life

* Complete response: total disappearance of all known disease for a period of at least four weeks.

** Partial response: in the case of bi-dimensionally measurable disease, a reduction of at least 50% in the sum of the products of the greatest perpendicular diameters measurable or of alI measurable lesions, determined through two observations not more than four weeks apart; in the case of uni-dimensionally measurable disease, reduction of at least 50% in the sum of the greatest diameters of all lesions determined through two observations not more than four weeks apart.

Search methods for identification of studies

All randomised clinical trials and controlled clinical trials on this subject were searched for electronically using MEDLINE (1966 to October 2013), EMBASE (1974 to October 2013), and the Cochrane Central Register of Controlled Trials (CENTRAL) (2012, Issue 3).

Electronic searches

The electronic searches are detailed in Appendix 1.

Searching other resources

1. Identification of further studies from references cited in the initial list of articles

2. We contacted pharmaceutical companies to identify any further unpublished or ongoing trials

3. A search of ongoing trials on the world-wide web: www.controlled-trials.com and www.cancer.gov

4. A search of grey literature on the world-wide web: http://cadth.ca/resources/grey-matters

Data collection and analysis

Selection of studies

Two authors independently assessed the abstracts of all references identified from the search strategy for inclusion and resolved any disagreements by discussion. We excluded abstracts referring exclusively to limited disease or treatments other than chemotherapy compared with placebo or BSC.

Data extraction and management

We entered the trials that met the selection criteria in a specific database. Foreign papers were translated when necessary to allow full assessment. Two authors independently extracted the data from the included studies using a previously standardised and tested form. We resolved discrepancies by discussion. Because most of the studies retrieved by the searches included both patients with limited and extensive disease, we sought specific information about patients with extensive disease from the principal investigators.

To assess the quality of each study we entered the following data in to the database.

  • Randomisation methods and allocation concealment.

  • Participants: description of patients, age, sex, and performance status.

  • Previous treatment.

  • Interventions: types of drugs and doses, cycles and frequency, other concomitant or sequential treatment, and time of follow-up.

  • Blinding strategies in the follow-up and assessment of the results.

  • Losses to follow-up and use or not of an intention-to-treat analysis.

  • Sample size and how this was determined.

  • Statistical analysis performed.

  • Baseline data.

  • Outcomes: survival time (medians or means at different periods and confidence intervals), complete and partial tumour responses, drug toxicity, and quality of life.

Assessment of risk of bias in included studies

We assessed bias in accordance with The Cochrane Collaboration's tool for assessing risk of bias (Higgins 2011). For each study we considered the following.

  • Selection bias: systematic differences between baseline characteristics of the comparison groups.

- Sequence generation: a description for allocating interventions to participants based on a random process.

* Low risk of bias: the authors describe a random component in the sequence generation process

* High risk of bias: the authors describe a non-random component in the sequence generation process

* Unclear: there is insufficient information about the sequence generation process to permit judgement of "Low risk" or "High risk"

- Allocation concealment: a description of any method to secure implementation of the random assignments to prevent that allocations could have been foreseen.

* Low risk of bias: the authors describe a central allocation, or sequentially numbered of identical appearance, or sealed envelopes opaque and numbered

* High risk of bias: there is a description of open random allocation, or unsealed or non-opaque envelopes, or alternation, or date of birth

* Unclear: there is insufficient information about the method of concealment to permit judgement of "Low risk" or "High risk"

  • Performance bias: systematic differences between groups in the care provided or in other factors different from the intervention.

-Blinding of participants and personnel: a description of measures used to blind study participants and personnel for knowledge of which intervention a participant received. Description of any evaluation of effective blinding was also considered. Blinding was considered important for the outcomes of tumour response, toxicity and quality of life, and less important for survival.

* Low risk of bias: blinding of participants and personnel and unlikely that could have been broken, or no blinding or incomplete blinding, but the review authors judge that the outcome is not likely to be influence by lack of blinding

*High risk of bias: blinding of participants and personnel and likely that the blinding could have been broken with the outcome likely to be influenced by lack of blinding, or no blinding or incomplete blinding, and the outcome is likely to be influenced by lack of blinding

* Unclear: there is insufficient information to permit judgement of "Low risk" or "High risk", or the study did not address blinding

-Blinding of outcome assessors: a description of measures used to blind the assessment of the outcomes

*Low risk of bias: blinding of the outcome assessment and unlikely that the blinding could have been broken, or no blinding but the review authors considered that the outcome measurement is not likely to be influence by lack of blinding

*High risk of bias: blinding of the outcome assessment and likely that the blinding could have been broken with the outcome measurement likely to be influenced by lack of blinding, or no blinding when the outcome assessment is likely to be influence by lack of blinding

*Unclear: there is insufficient information to judge the existence of "Low risk" or "High risk" or the study did not address this kind of blinding

  • Attrition bias: systematic differences between groups in withdrawals form the study.

- Incomplete outcome data: a description of the level of completeness of each outcome, including withdrawals, dropouts, and protocol deviations. Number of randomised patients in each group and number of evaluated patients with reasons for attrition reported.

*Low risk of bias: any one of the following. No missing outcome data; missing outcome data balanced in number across groups, with similar reasons for missing data; the proportion of missing outcomes compared with the observed event risk is not enough to have a clinically relevant impact on the intervention effect estimate

*High risk of bias: there is reason for missing outcome data likely to be related to the true outcome, with either imbalance in numbers or in reasons for missing data across groups; the proportion of missing outcomes compared with the observed event risk is enough to produce clinically relevant bias in the intervention effect estimate; "as-treated" analysis done with substantial departure of the intervention received form that assigned at randomisation.

*Unclear: there is insufficient information of attrition/exclusions to permit judgement of "Low risk" or "High risk"

  • Reporting bias: reporting a selection of a subset of the original outcome variables

-Selective reporting:

*Low risk of bias: any one of the following. The study protocol is available and all of the study's prespecified outcomes that are of interest in the review have been reported or the study protocol is not available but it is clear that the published report include all expected outcomes, including those that were prespecified

*High risk of bias: not all of the study's prespecified outcomes have been reported or one or more primary outcomes are reported using measurements, analysis methods or subsets of data that were not prespecified; or one or more reported primary outcomes were not prespecified, unless clear justification for their reporting is provided, such as an unexpected adverse effect, or one or more outcomes of interest in the review are reported incompletely so that they cannot be entered in a meta-analysis; the study report fails to include results for a key outcome that would be expected to have been reported for such a study

*Unclear: there is insufficient information to permit judgement of "Low risk" or "High risk"

For each domain, two authors independently made the judgment of risk of bias in accordance with the following classification: Low risk of bias; High risk of bias; Unclear.

Measures of treatment effect

Treatment effect measures had to be computed by means of relative risks (dichotomous data), hazard ratios, and standardised mean differences (continuous data). A narrative description of results was reported in the first-line chemotherapy studies. In the second-line chemotherapy studies it was possible to obtain an overall effect measure of two studies. The third study was reported in a narrative. The statistical analysis was done in line with the recommendations of The Cochrane Collaboration using RevMan 5.2. We pooled the study specific hazard ratio estimates and their standard errors to obtain an overall hazard ratio and its 95% confident interval. We used as the method of meta-analysis the inverse variance with the random-effects model of Der Simonian and Laird to calculate the effect size, because of the clinical and methodological heterogeneity of the studies. Heterogeneity in the effect estimates was quantified with the I2 statistic, which indicates the proportion of the total variation in the estimates that is caused by variation between studies not by chance. We considered I2 < 25% as representing low heterogeneity and I2 > 75% as representing high heterogeneity.

Missing data were dealt with by analysing only the available data and not if the trial authors did not provide them. The potential impact of missing data on the findings of the review were addressed in the discussion section.

Unit of analysis issues

The unit of analysis in all the included trials was the patient.

Dealing with missing data

To obtain data not available in the included studies, we attempted to contact the principal study investigators.

Assessment of heterogeneity

In terms of clinical heterogeneity, we explored any kind of variability among studies. We assessed baseline group imbalance as well as other sources of clinical heterogeneity such as different diagnostic methods among the studies and groups, adverse effects, differences in dose of drug, length of follow-up, and characteristics of participants. We explored heterogeneity in the meta-analysis by comparing studies in terms of methodological factors such as concealment of allocation, the definition and measurement of outcomes, as well as in relation to clinical heterogeneity.

Assessment of reporting biases

We took into account the following:

  • publication bias, by contacting pharmaceutical companies for non-published studies; duplicate publication; reporting of outcomes; no language restrictions for inclusion of trials;

  • the influence of external funding and commercial interests;

  • a search of multiple database sources, and consulting trials registries.

Since very few studies were included, we did not explore the use of funnel plots.

Subgroup analysis and investigation of heterogeneity

Subgroup analyses were to be done for subsets of participants (gender, age) or for subsets of studies (low risk versus high risk of bias) as a means of evaluating heterogeneous results, if there were a substantial number of studies.

As it was not possible to perform any subgroup analysis in this review, we reported either differences in study and patient characteristics, the way in which outcomes were defined or measured, protocol requirements, the length of follow-up, or differential losses to follow-up. That is, clinical or methodology heterogeneity.

Sensitivity analysis

No sensitivity analysis was performed due to the small number of included studies. References to better quality studies were described.

Results

Description of studies

Results of the search

From the literature search we obtained 16,282 references, 12,328 from MEDLINE and CENTRAL and 3954 from EMBASE, after excluding duplicates. We identified 50 potentially eligible trials in the first update and another five in the second update (Figure 1). Only five trials compared first or second-line chemotherapy versus placebo or BSC: two compared first-line chemotherapy versus supportive care or no active anticancer treatment (Kokron 1977b; Kokron 1982) and the other three compared second-line chemotherapy at relapse or progression versus symptomatic treatment or BSC (Ciuleanu 2010; O'Brien 2006; Spiro 1989). In the search for the second update we found the Ciuleanu trial; two reports of quality of life in relation to O'Brien 2006, one was a repetition of the same results of the study so it was excluded (see Characteristics of excluded studies), the other study (O'Brien 2007a) offered complementary information about quality of life and was included; another study, which was a retrospective report, was excluded. The Ciuleanu study was in abstract form and we consulted the Poniard website of the study (www.prnewswire.com/news-releases/poniard-announces-final-data-from-phase-3-spear-trial-of-picoplatin-in-small-cell-lung-cancer-95693244.html) for more complete information. We did not identify any studies of first-line chemotherapy versus BSC after 2003, when the original review took place. We received no response from the pharmaceutical companies that we contacted for information about any possible non-published studies except for Lilly and Astra Zeneca saying that they had no unpublished studies corresponding to our review. Contact was established with the principal author of the Ciuleanu 2010 study but he was not able to reply to our questions due to the fact that the study's data are the property of Poniard Pharmaceuticals; the company did not reply to his questions either.

Figure 1.

Study flow diagram.

Included studies

This review focused on the two trials by Kokron (Kokron 1977b; Kokron 1982), identified in the original review of 2003, and on the three trials by Spiro, O’Brien, and Ciuleanu (Ciuleanu 2010; O'Brien 2006; Spiro 1989), the only ones that fulfilled the inclusion criteria.

First-line chemotherapy versus best supportive care (BSC)

Two studies included a total of 88 patients. See Table 1 for further details.

Table 1. First-line chemotherapy versus best supportive care (BSC)
StudyGroup AGroup BGroup C
Kokron 1977bSupportive care: mainly antibiotics and analgesics

Ifosfamide monotherapy: 1500 mg/m2 ifosfamide

as short infusion iv. on day 1, followed by ifosfamide

1000 mg/m2 iv.on days 2 and 3. Repetition of ifosfamide

1000 mg/m2 iv on days 8, 9, 10 and 15,16,17.

Continuation of the therapy after six weeks with three day therapy (1000 mg/m2 daily each).

-------------------------------------------------
Kokron 1982No active anticancer treatment: infusion of Ringers solution 1000 ml + 1 gr vitamin C iv three times a week during an in-patient stay.Ifosfamide monotherapy: same schedule as the first study

Ifosfamide + CCNU (chloroethyl-cyclohexyl-nitrosourea):

ifosfamide therapy as described under Group B,

in addition to CCNU 60 mg/m2 on day 1 with repetition after six weeks.

Kokron 1977b included 34 patients (33 evaluable) selected from 1972 to 1974 with the objective of comparing ifosfamide with supportive care.

Kokron 1982 was published in 1982 and included 54 eligible patients (41 of them with extensive disease, 32 evaluable) selected between 1975 and 1979. The objective of this study was to compare ifosfamide versus ifosfamide plus chloroethyl-cyclohexyl-nitrosourea (CCNU) versus no active anticancer treatment. Both studies were carried out at the City Hospital Lainz, Vienna.

The inclusion criteria for the trials were: male patients; age under 70 years; inoperable small cell carcinoma of the bronchus with histological or cytological verification of tumour, or both; no previous tumour therapy; no large total tumour mass; no detected brain metastasis at the start of therapy; no chronic disease (cardiac, hepatocirrhosis, renal insufficiency, insulin dependent diabetes, or impediment in the region of the urinary tract) and good general condition (able to work) (see Characteristics of included studies, Figure 2, Figure 3).

Figure 2.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Second-line chemotherapy at relapse versus best supportive care (BSC)

Three studies included a total of 932 patients. See Table 2 for more information.

Table 2. Second-line chemotherapy at relapse versus best supportive care (BSC)
  1. *Patients with disease progression. First-line chemotherapy treatment consisted of either four or eight cycles of cyclophosphamide 1 g/m2 iv; vincristine 1.4 mg/m2 iv; on day 1, and etoposide capsules 100 mg orally eight hourly on days 1 to 3, every 21 days

    **Patients with relapse and who were unsuitable for further intravenous chemotherapy

    ***Patients with refractory disease or progression within six months after first-line platinum-based chemotherapy

    a = Best Supportive Care

StudyGroup AGroup B
Spiro 1989*Symptomatic treatment including palliative irradiation

Methotrexate 50 mg/m2 iv plus doxorubicin 50 mg/m2 iv, repeated every

three weeks with a maximum of nine courses

O'Brien 2006**BSCa

Topotecan in capsules containing the equivalent to 0.25 mg, doses

2.3 mg/m2 on days 1 to 5 every 21 days, at least four cycles + BSC.

Ciuleanu 2010***BSCaPicoplatin 150 mg/m2, every 3 weeks median cycles 3

Spiro 1989 included 616 patients (610 evaluable), 414 of them with advanced SCLC, recruited from 1982 to 1985 at various institutions in the United Kingdom. Second-line chemotherapy or symptomatic treatment was administered to 390 patients in progression (170 evaluable in active treatment).

O'Brien 2006 included 141 patients (141 evaluable) recruited from 2000 to 2004 at 40 centres in Europe, Canada, and Russia.

Ciuleanu 2010 included 401 patients (321 evaluable) with refractory disease or SCLC in progression within six months of first-line chemotherapy, recruited from international multi-centre sites.

The inclusion criteria for the trials were: both sexes; age under 75 years (Spiro 1989) or ≥ 18 years (O'Brien 2006); median age 58 years (Ciuleanu 2010); SCLC diagnosed by histology or by cytology; previous chemotherapy; no symptomatic brain metastases (O'Brien 2006); no severe co-morbidities; no previous treatment with topotecan (O'Brien 2006); any performance status (Spiro 1989) or between 0 and 2 (Ciuleanu 2010; O'Brien 2006).

The primary endpoint in the five trials was overall survival. Secondary endpoints were response rate, quality of life, and safety.

Supplementary data from the Kokron studies were supplied by the author. We were unable to establish contact with the rest of authors.

Excluded studies

A total of 48 studies were excluded. These studies did not include a comparison of a chemotherapeutic agent with placebo or BSC, were phase II trials without a control group, or they compared second-line chemotherapy in a complete or partial response with observation (see Characteristics of excluded studies).

Risk of bias in included studies

See the 'Risk of bias' tables in Characteristics of included studies.

Allocation

The allocation strategy was considered unclear for the five trials (Ciuleanu 2010; Kokron 1977b; Kokron 1982; O'Brien 2006; O'Brien 2007a; Spiro 1989), while allocation concealment was judged adequate for one trial (O'Brien 2006; O'Brien 2007a) and unclear for the other four trials (Ciuleanu 2010; Kokron 1977b; Kokron 1982; Spiro 1989). Although we actively contacted the authors requesting information about their trials, we received no replies.

Blinding

Blinding was not used in four of the five trials but the main outcome of survival is not likely to be influenced by lack of blinding.

Incomplete outcome data

Losses to follow-up and their reasons were reported in only one trial (O'Brien 2006; O'Brien 2007a). Attrition data seemed to be different between groups.

Selective reporting

Relevant outcomes for this review were reported in only one trial (O'Brien 2006; O'Brien 2007a).

Other potential sources of bias

Four studies (Ciuleanu 2010; Kokron 1977b; Kokron 1982; Spiro 1989) were considered to be affected by other sources of bias.

Effects of interventions

See: Summary of findings for the main comparison First-line chemotherapy compared with BSC for extensive SCLC; Summary of findings 2 Second-line chemotherapy at relapse or progression in extensive SCLC

A) First-line chemotherapy versus best supportive care (BSC)

The results were based on a total of 65 assessed patients with advanced disease who were included in two studies (Kokron 1977b; Kokron 1982). Information about the baseline data of the patients in the two arms and a calculation of sample size were not found in the Kokron reports. The published studies of Kokron did not provide sufficient details about the analyses performed, although survival curves were used to compare survival between groups in both studies. Information that was later provided by the author referred to ANOVA to compare the three groups in the second study (Kokron 1982).

Because it was not possible to obtain data suitable for a combined analysis, the results were described separately for each study according to the data provided by the author. Results for each study are presented in Table 3.

Table 3. Results of the studies included in the review (first-line chemotherapy versus best supportive care (BSC)
  1. MST = mean survival time; PR = partial response; CR = complete response; CCNU = cyclohexylnitrosourea

Study

ID

Objective Survival Tumour response Toxicity
Kokron 1977bTo compare supportive care vs IfosfamideMST1: supportive care group = 93 days vs 172 Ifosfamide group (range:11 to 544 vs 1 to 1096)

PR2 = 0%  supportive care group vs 47% Ifosfamide group

CR3= 0%

Haematological: 0%  supportive care group vs  94.1% Ifosfamide group
Kokron 1982To compare no active anticancer treatment vs ifosfamide vs Ifosfamide + CCNUMST: in extensive disease: no active anticancer treatment = 56 days vs 134 Ifosfamide vs 122 Ifosfamide + CCNU. Grouped data: no active anticancer treatment, median = 161.5 (MST = 208.4), Ifosfamide, median = 341 (MST = 348.5), Ifosfamide + CCNU, median = 252 (MST = 322.3)Grouped data (extensive and limited disease): CR = 1.9% Ifosfamide; PR = 16.9 Ifosfamide, and 13.2% Ifosfamide + CCNUGrouped data: leucopenia:0% no active anticancer treatment; 15% Ifosfamide; 15.8% Ifosfamide + CCNU. Vomiting and hair loss: 0% no active anticancer treatment, 70% Ifosfamide; 68.4% Ifosfamide + CCNU. Other: 0% no active anticancer treatment; 55% Ifosfamide; 52.6% Ifosfamide + CCNU
Survival
  • Mean survival

 In Kokron 1977b, treatment with ifosfamide gave a mean survival time that was 79 days longer than in the supportive care group. Available data did not allow calculation of the confidence interval.

In Kokron 1982, the data for all patients (limited and extensive disease) showed a significant difference (P = 0.01) in the mean survival time between the no active anticancer treatment and the ifosfamide groups and between the no active anticancer treatment and the ifosfamide + CCNU groups. No significant differences were found between the ifosfamide and ifosfamide + CCNU groups. For patients with extensive disease, where only the mean survival time was reported by the author, the mean survival time was 78 days longer (P = 0.01) for the ifosfamide group and 66 days longer (P = 0.01) for the ifosfamide + CCNU group in comparison with the no active anticancer treatment group. In addition, in the ifosfamide group the mean survival time was 12 days longer (P = 0.05) than in the group receiving ifosfamide + CCNU. The quality of the evidence was very low for this outcome. (See Summary of findings for the main comparison).

Tumour response

No patients in Kokron 1977b showed a complete response. However, eight patients (47%) in the ifosfamide group showed a partial response at 5.6 months versus no patients in the control group. In Kokron 1982 only one patient in the ifosfamide treatment group had a complete tumour response. Nine patients in the ifosfamide group and seven in the ifosfamide + CCNU group showed a partial tumour response. These data referred to all patients (limited and extensive disease) as the author did not report data specifically for patients with extensive disease.

Toxicity

In Kokron 1977b six patients (94.1%) treated with ifosfamide had haematological toxicity (its type was not reported) compared with none in the supportive care group. In Kokron 1982, data on toxicity referred to all patients (limited and extensive disease), where no cases of toxicity were recorded for the no active anticancer treatment group (very low-quality evidence).

Quality of life

No validated instruments were used to assess quality of life in the Kokron studies (Kokron 1977b; Kokron 1982). Patients’ levels of activity were assessed (Kokron 1982) using a simple scale: 0: no symptoms, 1: subjective symptoms only, 2: objective signs of illness but able to work, 3: needing care, 4: bedridden, 5: confused and disorientated. No information or reference was given concerning its validation. Data based on this scale were reported only for the beginning of treatment (for all patients together, limited and extensive disease) when 34 had an activity level of 1 and 18 were level 2. The data also showed that 61% of patients in the first group, 80% in the second, and 53% in the third presented a subjective feeling of disease with a score of 1 on the scale. No specific information was mentioned for patients with extensive disease and no measurement of this scale was reported at the end of the study.

B) Second-line chemotherapy versus best supportive care (BSC) at relapse or progression

The results of this comparison were based on a total of 932 patients with advanced disease who were included in three studies (Ciuleanu 2010; O'Brien 2006; Spiro 1989). All patients in progression or relapse were considered as having extensive disease in this review. Information about the baseline data of patients in the two arms and a calculation of sample size were described in O'Brien 2006.

In Spiro 1989, randomisation took place at diagnosis not relapse and patient characteristics were described at presentation with either limited or extensive disease. O'Brien provided baseline characteristics of the patients but Ciuleanu did not.

Patients in Spiro 1989 were allocated to one of two groups: 192 (59 women) to the methotrexate-doxorubicin group and 198 (63 women) to the symptomatic treatment group. Patients in O'Brien 2006 were allocated to one of two groups: 71 (19 women) to the topotecan group and 70 (19 women) to the BSC group. Patients in the Ciuleanu 2010 study were allocated to one of two groups: 268 picoplatin, 133 BSC. Results for each study are presented in Table 4 and in the 'Data analysis' section.

Table 4. Results of the studies included in the review (second-line chemotherapy versus best supportive care (BSC)
  1. MST = median survival time; PR = partial response CR = complete response; BSC = best support care; CI = confidence interval; EQ-5D = EuroQol-5-Dimensions (mobility, self-care, usual activities, pain and discomfort, anxiety and depression) with rating 1 (no problem) to 3 (extreme problem); PSA = Patient Self Assessment questionnaire, with rating 1 (not at all) to 4 (very much)

Study

ID

Objective Survival Tumour response Toxicity Quality of life
Spiro 1989To assess the effects of duration of chemotherapy on survival, and at relapse, the effects of further chemotherapy compared with symptomatic treatmentMST1: symptomatic = 77 to 84 days vs 140 to 105 days methotrexate-doxorubicin group (in previous chemotherapy: 4 cycles-8 cycles, vs 4 cycles-8 cycles) Only significant 77 vs 140 days (P < 0.001)PR2 or CR3 in methotrexate-doxorubicin group = 22.3%Not provided for second-line chemotherapy-------------------
O'Brien 2006To define risk and benefit of second-line chemotherapy at relapse vs BSC in patients unsuitable for standard intravenous chemotherapy

MST: in BSC 97.3 days (95% CI, 77.7 to 130.2) vs  181.3 days (95% CI, 128.1 to 221.2) in Topotecan group (log-rank P = 0.01)

Subgroup analysis

a) treatment-free interval <=60 days of first-line chemotherapy: topotecan 163.1 (95% CI 74.9 to 216.3) vs BSC 92.4 (95% CI 49 to 147)

b) performance status 2: topotecan 146.3 (93.8 to 188.3) vs BSC 53.9 (37.1 to 91.7)

c) female: HR 0.38 (95% CI 0.18 to 0.76)

d) liver metastasis: HR 0.58 (95% CI 0.38 to 0.88)

PR  in the topotecan group = 7% (95% CI, 2.33 to 15.67)

CR 0%

In treatment-free interval <=60 days: PR 18% (5.2 to 40.3) vs 2% 80.05 to 10.9) in treatment-free interval >60 days

Topotecan group: Haematological: neutropenia ¾, 61%; thrombocytopenia ¾, 38%, anaemia ¾, 25%. Nonsepsis infection>= grade 2, 14%, vs 12% in BSC. Sepsis 4% in topotecan vs 1% in BSC.

Non-haematological 3/4 in topotecan vs BSC: dyspnoea 3% vs 9%; fatigue 4% vs 4%.

Toxic deaths in topotecan: 6%

all-cause mortality within 30 days: 7% topotecan vs 13% BSC

Difference in rate of deterioration per 3-month intervals in EQ-5D was 0.15 (95% CI 0.05 to 0.25)

In PSA questionnaire:

shortness of breath RR 2.18 (1.09 to 4.38)

interference with sleep 2.16 (1.15 to 4.06)

fatigue 2.29 (1.25 to 4.19)

All in favour of topotecan

Ciuleanu 2010To assess survival of patients treated with second-line chemotherapy who were refractory or who progressed within six months of first-line platinum-based chemotherapy

MST: in BSC 138 days (95% CI 112 to 168)

in picoplatin 144 days (95% CI 133 to 175)

HR: 0.817 95% CI 0.65 to 1.03 (P=0.00895)

Time to progression:79 days picoplatin, 47 days BSC

HR: 0.610 (P=0.0002)

Median progression-free survival: 63 days picoplatin, 46 days BSC

HR: 0.783 (P=0.0281)

Subgroup analysis

a) patients who did not receive chemotherapy post-study (273: 194 picoplatin vs 79 BSC): picoplatin 128 days (95% CI 112 to 140), BSC 101 (77 to 140), HR 0.730 (P=0.0345)

b) patients refractory or relapsed within 45 days of first-line chemotherapy (294: 202 picoplatin vs 92 BSC):

picoplatin 149, BSC 129,

HR 0.717 (P=0.0173)

No data provided

Picoplatin group: Haematological: 3/4 in >10%:

44% thrombocytopenia

29% anaemia

18% neutropenia

<1% febrile neutropenia

No deaths to haematological toxicity, no bleeding

Nonhaematological:

0.8% grade 3 neuropathy

11% asthenia

-------------------------
Survival
  • Median survival

In Spiro 1989, survival data were provided in accordance with the first-line chemotherapy group allocation: treatment with methotrexate-doxorubicin at relapse gave a median survival time of 63 days longer than in the symptomatic treatment group (P < 0.001) for patients allocated to receive four cycles of first-line chemotherapy, and 21 days longer (P = 0.160) for patients allocated to receive eight cycles of first-line chemotherapy. Confidence intervals were not reported. No survival comparisons between the symptomatic group versus the methotrexate group were provided.

In O'Brien 2006, treatment with topotecan at relapse gave a median survival time 84 days longer than in the BSC group (log-rank P = 0.01). The hazard ratio for overall survival, adjusted for the stratification factors of sex, performance status 0/1 or 2, treatment-free interval ≤ 60 days or > 60 days, presence of liver metastases, was 0.61 (95% CI 0.43 to 0.87).

In all subgroup analyses in the O'Brien 2006 study (treatment-free interval ≤ 60 days after first-line chemotherapy, performance status 2, female, liver metastasis) (see Table 4) the survival differences were maintained.

In Ciuleanu 2010, treatment with picoplatin in refractory patients or for patients in progression gave a median survival time six days longer than in the BSC group (HR 0.82, 95% CI 0.65 to 1.03, P = 0.0895). The time to progression was significant longer (32 days) in the picoplatin group in comparison with the BSC group, and the median progression-free survival was also significant longer (by 17 days) in the picoplatin group. Patients of both groups received second-line chemotherapy post-study, 27.6% in the picoplatin arm and 40.6% in the BSC arm (P = 0.012).

In the first subgroup analysis of patients who did not receive second-line chemotherapy post-study in either arm (picoplatin or BSC), and in the second subgroup analysis of patients refractory or relapsed within 45 days of first-line chemotherapy, a significant longer survival time was observed in the picoplatin group compared to the BSC group (see Table 4).

The overall survival measured by means of the combined hazard ratio for topotecan plus picoplatin was 0.73 (95% CI 0.55 to 0.96, P = 0.03; low-quality evidence), with medium heterogeneity (47%), favouring second-line chemotherapy over BSC (see Summary of findings 2, Analysis 1.1, Figure 4).

Figure 4.

Forest plot of comparison: 1 Second-line chemotherapy at relapse versus BSC, outcome: 1.1 Overall survival.

  • Survival at six months

Data were extracted from the survival curves and in accordance with the previous chemotherapy received: Spiro 1989 symptomatic group previously four cycles 20%; previously eight cycles 22%; in the methotrexate-doxorubicin group previously four cycles 37%, previously eight cycles 32%.

Six-month survival rates were 26% in the BSC group and 49% in the topotecan group (O'Brien 2006). No data were provided in the picoplatin study (Ciuleanu 2010).

  • Survival at one year

Data were extracted from the survival curves and in accordance with previous chemotherapy received: Spiro 1989 symptomatic group previously four cycles 4%, previously eight cycles 2%; in the methotrexate-doxorubicin group previously four cycles 6%, previously eight cycles 5%.

One-year survival rates (extrapolated from the survival graphs) were 13% in the BSC group and 19% in the topotecan group (O'Brien 2006). No data were provided in the picoplatin study (Ciuleanu 2010).

  • Survival at two years

Two-year survival in patients with extensive disease at presentation, for both groups in Spiro 1989, were shown according to first-line chemotherapy allocation: 2% to 2.7% for the symptomatic group allocated to four and eight cycles respectively, and 1.1% to 1% for the methotrexate-doxorubicin group allocated to four and eight cycles, respectively. In O'Brien 2006 the two-year survival rates (extrapolated from the survival graphs) were 3% and 9% in the BSC and topotecan groups, respectively. No data were provided in the picoplatin study (Ciuleanu 2010).

Tumour response

In Spiro 1989, in the methotrexate-doxorubicin group the partial or complete response was 22.3%; in O'Brien 2006 in the topotecan group there was no complete response and 7% showed a partial response; in Ciuleanu 2010 no data were provided for picoplatin.

Toxicity

No data were provided for the methotrexate-doxorubicin group (Spiro 1989). Topotecan showed more toxicity than BSC (Table 4), with 6% toxic deaths in O'Brien 2006. In the Ciuleanu 2010 study picoplatin showed more toxicity than BSC but with no deaths due to haematological toxicity (moderate-quality evidence).

Quality of life

Quality of life was not measured for patients receiving methotrexate-doxorubicin chemotherapy in Spiro 1989 or for patients receiving picoplatin at relapse in Ciuleanu 2010.

O'Brien 2006 used two questionnaires: the Patient Self Assessment (PSA) questionnaire, which resembles the Lung Cancer Symptom Scale, evaluating the degree to which patients experienced nine common symptoms on a Likert scale for severity (from 1, not at all, to 4, very much); and the EuroQol-5 Dimensions Health questionnaire (EQ-5D), consisting of the five health status dimensions mobility, self-care, usual activities, pain and discomfort, and anxiety and depression. The questionnaire has a rating scale from 1 (no problem) to 3 (an extreme problem), and a visual analogue scale ranking from 0 (worst imaginable health state) to 100 (best imaginable health state). Patients in both arms were evaluated before each course of second-line chemotherapy, or at each visit in the BSC arm. Not all patients completed the pre-specified questionnaires at the different visits.

The difference in rate of deterioration in the score at three-month intervals in the EQ-5D questionnaire was significantly lower in the topotecan group (0.15, 95% CI 0.05 to 0.25). The PSA questionnaire, conducted in 61 (86%) patients in the topotecan group and 48 (69%) in the BSC group, gave significant risk ratios for shortness of breath (RR 2.18, 95% CI 1.09 to 4.38), interference with sleep (RR 2.16, 95% CI 1.15 to 4.06) and fatigue (RR 2.29, 95% CI 1.25 to 4.19) (see Table 4).

Patients with a lower functional status based on their Eastern Cooperative Oncology Group (ECOG) performance status (O'Brien 2007a) reported more severe symptoms in the PSA questionnaire: mean score 12.31 for ECOG 0, 16.37 for ECOG 1, and 17.58 for ECOG 2 (P = 0.0002). Mean changes from baseline in the questionnaire score showed an association between tumour response and symptom control (P = 0.0016). Patients with a partial response (n = 5) had a baseline score of 15.07, while the mean score on treatment reduced to 12.64 representing a change from baseline of -2.43, or 16.12% improvement in symptoms. Patients with stable disease (n = 31) did not change their score between baseline and on treatment (15.33 versus 15.35) (low-quality evidence).

Discussion

The present study involved a comprehensive search of electronic databases and contact with authors and companies in order to identify trials that evaluated the efficacy of first-line and second-line chemotherapy at relapse as compared to best supportive care (BSC) or no active anticancer treatment.

Summary of main results

Only the two studies by Kokron (Kokron 1977b; Kokron 1982) met the inclusion criteria for the review of first-line chemotherapy compared with supportive care. These studies showed that for patients with extensive disease, ifosfamide gives an increased survival benefit (around 80 additional days) over supportive treatment but this benefit is not increased by the addition of CCNU. A partial tumour response of 47% was shown in the group on active therapy in the first study. Adverse effects were present in the chemotherapy group.

The results of this review in relation to second-line chemotherapy on relapse or at progression showed an increased survival benefit of between 63 days for patients receiving methotrexate-doxorubicin and 84 days for patients receiving topotecan over supportive treatment. No survival benefit was shown with picoplatin over BSC. According to Spiro 1989, the increase in survival after progression was significant in those patients who had previously received four cycles of first-line chemotherapy. The pooled analysis of topotecan and picoplatin gave an HR of 0.73 (95% CI 0.55 to 0.96) and significant improvement in overall survival. O'Brien 2006 reported that the increase in survival was preserved in women, the subgroup of patients with a treatment-free interval ≤ 60 days, patients with a performance status of 2, and in patients with liver metastases. This survival benefit was accompanied by an improvement in the quality of life, with a difference in the rate of deterioration in EQ-5D questionnaire score at three monthly intervals significantly in favour of the topotecan group, together with an improvement in shortness of breath, interference with sleep, and fatigue in the PSA questionnaire. Tumour response (partial or complete) in the methotrexate-doxorubicin group was 22.3%. Five patients (7%, 95% CI 2.33 to 15.67) showed a partial response in the topotecan group without any complete responses reported.

Overall completeness and applicability of evidence

The evidence that is available on the effectiveness and toxicity of first-line chemotherapy for patients with extensive SCLC comes from two studies that compared ifosfamide and ifosfamide + CCNU versus supportive treatment (Kokron 1977b; Kokron 1982). It remains unclear whether the results would be the same in women, patients with poor performance status, and in patients older than 70 years. Toxic deaths due to chemotherapy as well as the balance between quality of life and life expectancy remain unclear. At the time the studies were conducted no validated instruments for measuring quality of life were available. Such instruments were first developed in the 1980s, mainly with the objective of measuring self-perceived health in patients when the efficacy of a treatment was being evaluated (Lacombe 1997).

Despite the methodological limitations of these trials, carried out more than 25 years ago, they seem to have provided some evidence to support Green's hypothesis (Green 1969) that first-line chemotherapy treatment for patients with advanced SCLC produces benefits in terms of more prolonged survival. Although a large number of clinical trials comparing different regimens have been conducted in advanced SCLC, to date no consistent findings have indicated what is the most effective combination. Platinum-based combination chemotherapy regimens have increased complete response rates with no significant benefit in survival when compared to non-platinum chemotherapy regimens (Amarasena 2009), and at present are the standard first line treatment for patients with SCLC.

The evidence available on the effectiveness of second-line chemotherapy at relapse or progression in patients with extensive SCLC comes from three studies that compared methotrexate-doxorubicin, topotecan, and picoplatin versus symptomatic treatment or BSC (Ciuleanu 2010; O'Brien 2006; Spiro 1989). It remains unclear whether the results would be the same in females, patients with treatment-free interval ≤ 60 days after first-line chemotherapy, and in patients with poor performance status. The O'Brien 2006 study showed an improvement in survival in these subgroups and no subgroup-treatment effect interaction was reported. However, despite the fact that the trial was powered to determine the overall effect of treatment, virtually all subgroup analyses were underpowered. If a genuine subgroup-treatment effect interaction exists, the chance of a false negative result with a statistical test of interaction will be greater than the 5% false positive rate in a trial in which no true interaction exists (Rothwell 2005). Therefore, interpretation of the risks and benefits of topotecan in the subgroups should be done cautiously, and the findings need validation by replication of the data.

The relationship between the first-line chemotherapy received and the effect of second-line chemotherapy remains unclear. The most commonly received first-line chemotherapeutic agents in O'Brien 2006 and Ciuleanu 2010 were cisplatin or carboplatin and etoposide, consistent with current guidelines. It has been suggested that the clinical benefit of re-treatment correlates with the type of first-line chemotherapy and treatment characteristics at the time of relapse (treatment-free period), a presumption that is difficult to clarify in this study given the homogeneity of the type of first-line chemotherapy administered to patients.

Although treatment with picoplatin showed an improvement in the median time to progression and in the median progression-free survival, both are surrogates outcomes for the final outcome or overall survival. A recent review (Ciani 2013) showed that, on average, trials using surrogate outcomes reported treatment effects that were 28% to 48% higher than trials using final relevant outcomes for patients. Surrogate trials were twice as likely to report positive treatment effects compared with the final outcome trials, and the findings were not explained by differences in risk of bias or other trial characteristics.

The quality of life of patients receiving second-line chemotherapy remains unclear. In this review only the topotecan study (O'Brien 2006) addressed this question, with an improvement in shortness of breath, quality of sleep, and fatigue.

In summary, the trials included in the review are clinically diverse and few; one of them has not been published in a peer-reviewed journal (Ciuleanu 2010), the survival benefit is limited, a few days, and the adverse effects are present in considerable proportions. Therefore, the evidence on the benefits of second-line chemotherapy in patients with extensive SCLC at relapse or progression is scarce, making it difficult to generate recommendations to clinicians and patients.

Quality of the evidence

The quality of the evidence available for all comparisons was moderate to very low, mainly because there were only two studies for first-line chemotherapy and three for second-line chemotherapy, with small sample sizes and significant risk of bias. Selection bias was considered unclear in all studies except O'Brien 2006. Performance bias in relation to survival seemed to be at low risk of bias in all studies, although at unclear risk in relation to toxicity and quality of life. Detection bias in relation to survival was considered to be at low risk in all studies but unclear in relation to tumour response, toxicity and quality of life. Attrition bias in relation to survival was considered to be at low risk in O'Brien 2006 and unclear in the rest of studies. In relation to tumour response, toxicity and quality of life attrition bias was at a mix of low, unclear and high risk bias. Selective reporting was considered to have unclear bias in all studies except O'Brien 2006, and four studies were considered to be affected by other bias.

For first-line chemotherapy the quality of the evidence for the overall survival and adverse events was very low. No evidence was found for quality of life. For second-line chemotherapy the quality of the evidence for the overall survival and quality of life was low, and moderate for toxic deaths and adverse effects.

In relation to other bias, the lack of information on the baseline characteristics of the treatment groups in the Kokron studies, Spiro, and Ciuleanu means that the presence of uncontrolled confounding variables cannot be excluded. Moreover, the small number of patients in each comparison group in the Kokron studies may have led to an overestimation of the effects, as has been described in other circumstances (Moore 1998). The ANOVA statistical analysis in the Kokron studies is inadequate to compare mean survival between groups. As well, the Kaplan-Meier method was inadequate and the statistical analysis should have included a Mantel-Haenszel analysis to compare treatment groups (Kleinbaum 1996). Two trials (Ciuleanu 2010; O'Brien 2006) were designed by the sponsors Poniard Pharmaceuticals and Glaxo-Smith-Kline, respectively, who held and analysed the data. Eight of the 10 authors in the O'Brien study disclosed potential conflicts of interest, all of them with the same sponsor, and the trial favoured the drug. A recent systematic review (Lundh 2012) on the relationship between pharmaceutical industry sponsorship and research outcomes found that clinical trails sponsored by the pharmaceutical companies are significantly more likely than non-commercially funded studies to report favourable efficacy and safety results and conclusions. The findings were consistent across a wide range of disease states and drugs, regardless of the type of research being assessed. The review did not find a difference in risk of bias of drugs and device trials sponsored by industry compared with non-industry sponsored trials. However, industry sponsored trials more often had low risk of bias from blinding. Although the research methods of trials sponsored by drug companies are reported to be at least as good as or, in many cases, even better than those of non-industry funded research, this does not guarantee the absence of bias.

Except for the Kokron 1982 blinding was not used to evaluate the results so measurement bias for variables such as tumour response, toxicity, and quality of life cannot be excluded, although this is less important when survival is the outcome being measured (Day 2000).

The attrition bias shown in the studies prevents a full intention-to-treat analysis being carried out (Dumville 2006). No information on the cause of the missing data is reported, neither were there analytical methods to deal with data that are not missing at random (Bell 2013).

The major strength of the review is that it is a comprehensive analysis of all the studies undertaken on the subject. The major limitation is the scarce number of studies included, the unclear risk of bias in the first-line chemotherapy studies and the majority of second-line chemotherapy studies, except for only one study that was judged to be at low risk of bias in some aspects such as allocation concealment, performance and detection of survival, attrition and outcome reporting (O'Brien 2006). Given the paucity of available robust studies assessing the clinical effects of first-line chemotherapy and second-line chemotherapy at relapse or progression over BSC, the evidence is insufficient to judge the effectiveness and safety of chemotherapy for the treatment of extensive SCLC with the desired degree of precision.

Potential biases in the review process

There was no language restriction in the search for trials and trial selection. Pharmaceutical companies were contacted for supplementary data on any non-published studies but no answer was received so publication bias could not be excluded. Conference abstracts were not searched. Given the few trials included in the review, no funnel plot could be drawn.

Agreements and disagreements with other studies or reviews

We did not find any systematic review of first-line chemotherapy treatment in advanced SCLC to compare with the results of this review with. In spite of limited evidence concerning the benefits of first-line chemotherapy, many would now consider it unethical not to give chemotherapy to patients with advanced SCLC.

According to a few studies, it seems that patients with advanced disease and poor prognostic factors (brain metastases, low performance status, biochemical anomalies such as increased lactate dehydrogenase (LDH), alkaline phosphatase, and sodium anomalies) have a higher mortality during the first chemotherapy cycle as a consequence of toxicity (Bernhard 1996; Rawson 1990; Souhami 1988). None of the studies included in this review reported on the number of early deaths due to chemotherapy in patients with a poor prognosis. So the benefits of first-line chemotherapy in patients with poor prognosis remain unclear.

We did find four reviews of second-line chemotherapy in SCLC patients at relapse or progression to compare with the results of this review. The Seidenfeld 2006 evidence report made reference to two studies included in this review with the same conclusions as in our review. Cheng 2007 included only the trial of O'Brien, with results similar to our review. Riemsma 2010 included the O'Brien study and six other studies outside the inclusion criteria of this review, with similar findings to our review. Loveman 2010 developed an economic model to estimate the cost-effectiveness of oral or intravenous topotecan compared with BSC for patients with relapsed SCLC for whom re-treatment with the first-line regimen was not considered appropriate. It included the O'Brien study and four other trials outside the scope of this review. The cost difference gave an incremental cost-effectiveness ratio (ICER) of GBP 26,833 per QALY gained.

Benefit of first-line chemotherapy was shown in Sundstr¢m 2005, an observational study that reported a survival benefit in favour of the chemotherapeutic group in comparison with supportive care (5.3 and 2.2 months, respectively).

However, no survival benefit was found for second-line chemotherapy in the retrospective Kim 2008 study in which disease recurrence was more than three months compared with disease recurrence within three months without the chemotherapy.

The unclear benefit of second-line chemotherapy in women that was found in this review with topotecan has been remarked on by Weatley-Price 2010 in a pooled analysis of chemotherapy trials, although at the same time the women experienced greater levels of grade 3/4 emesis.

The treatments in the studies of this review were administered intravenously and thus it was not possible to evaluate the benefit of oral chemotherapy over BSC. Yet oral administration is preferred by patients over intravenous therapies, including in the case of topotecan (Gralla 2004).

The effectiveness of intense follow-up after a response in patients with SCLC treated with first-line chemotherapy has not been addressed in this review. However, an evaluation study (Sugiyama 2008) reported that survival was significantly longer in an intense follow-up arm (MD 9 months, 95% CI 7 to 12 months) than in the non-intensive arm (4 months, 95% CI 2 to 7 months).

Given the poor prognosis of extensive SCLC and the report that 69% of patients with extensive lung cancer have inaccurate expectations about the benefits of chemotherapy that are not consistent with their life expectancy (Weeks 2012), this misunderstanding could represent an obstacle in optimal end-of-life planning and care.

Authors' conclusions

Implications for practice

SCLC is an aggressive and fatal disease in the short term. Two small RCTs from the 1970s suggest that first-line chemotherapy treatment (based on ifosfamide) may provide a small survival benefit (less than three months) in comparison with supportive care or placebo infusion in patients with advanced SCLC, particularly those younger than 70 years with good performance status. The benefits remain unclear in older patients, women, and patients with poor performance status. However platinum-based combination chemotherapy regimens have been shown to increase complete response rates when compared to non-platinum chemotherapy regimens with no significant difference in survival, and so these are currently the standard first-line treatment for patients with SCLC.

Second-line chemotherapy (based on topotecan) at relapse or on progression could increase survival by a few weeks in patients with no symptomatic brain metastases and any performance status compared to best supportive care (BSC). Its benefits remains unclear in women, patients with a treatment-free interval ≤60 days after first-line chemotherapy and in patients with liver metastases.

Globally, the evidence on which these conclusions are based is very scarce and of uncertain or low quality, so it would be necessary to explain to patients the weak evidence for any benefit attributable to chemotherapy as well the poor prognosis of the disease in the short term, in a way that is understandable by them, in order to avoid inaccurate expectations.

Implications for research

Since chemotherapy with more modern agents is now established treatment for patients with SCLC, it is likely that a new study comparing active chemotherapy with BSC would be considered unethical. New studies comparing different drugs or combinations could be undertaken, with ifosfamide as the necessary comparator. The feasibility and ethics of conducting studies comparing active treatment with BSC in patients with poor prognosis (with poor performance status, women, or older people) could also be considered. This would allow the setting of more clear and explicit criteria to guide when to stop chemotherapy, with the objective of preserving quality of life.

In patients with SCLC that has relapsed or progressed, the uncertainty about the trade-offs between potential benefits (a gain of some weeks in survival), probable toxicity, and associated costs, coupled with the paucity and dubious quality of the available studies, would justify more trials comparing second-line chemotherapy versus BSC. These randomised clinical trials should be well-designed and large enough to allow assessment of the benefits in pre-specified subgroups such as women, older patients, people with low performance status, and previous response to first-line chemotherapy. These trials should also evaluate the effectiveness of follow-up schedules after first-line chemotherapy and include the use of validated instruments to measure quality of life. In addition, the information provided by the new, studies could be presented to patients and relatives in some objective way to enable them to make well-informed decisions on the potential risks and benefits of available therapeutic options.

Acknowledgements

We would Iike to thank Dr Kokron for providing further information on his trials. We are grateful to Dr Elinor Thompson for helpful comments provided during the development of the initial review and to Marta Roque for her helpful comments provided in the analysis of the study data. Thanks to Desiree West (consumer of the Lung Cancer Group) for her comments on the review.

Data and analyses

Download statistical data

Comparison 1. Second-line chemotherapy at relapse versus best supportive care (BSC)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Overall survival2 Hazard Ratio (Random, 95% CI)0.73 [0.55, 0.96]
Analysis 1.1.

Comparison 1 Second-line chemotherapy at relapse versus best supportive care (BSC), Outcome 1 Overall survival.

Appendices

Appendix 1. Electronic searches

1.1 MEDLINE search using the following strategy (Ovid)

  1. small cell lung cancer. mp

  2. oat cell.tw

  3. SCLC.tw

  4. 1 or 2 or 3

  5. randomized controlled trial.pt

  6. controlled clinical trial.pt

  7. randomized controlled trials/

  8. random allocation/

  9. double-blind method/

  10. single-blind method/

  11. or/5-10

  12. clinical trial.pt

  13. exp clinical trial/

  14. (clin$ adj25 trial$).ti,ab

  15. ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab

  16. placebos/

  17. placebo$.ti,ab

  18. Random$.ti,ab

  19. research design/

  20. or/12-19

  21. comparative studies/

  22. exp evaluation studies/

  23. follow-up studies/

  24. prospective studies/

  25. (control$ or prospectiv$).ti,ab

  26. or/21-25

  27. 11 or 20 or 26

  28. 4 and 27

  29. limit 28 to human

1.2 EMBASE search using similar strategy

1. small cell lung cancer.mp
2. oat cell.tw.
3. SCLC.tw
4. lung small cell cancer DE
5. 1 or 2 or 3 or 4
6. random$.ti,ab

7. factorial$.ti,ab

8. placebo$.ti,ab

9. (doubl$ adj blind$).ti,ab

10.singl$ adj blind$).ti,ab

11. assign$.ti,ab

12. allocate$.ti,ab

13. DOUBLE-BLIND PROCEDURE.sh

14. RANDOMIZED CONTROLLED TRIAL.sh

15. SINGLE-BLIND PROCEDURE.sh

16. or/6-15

17. symptomatic therapy only

18. palliative cancer therapy.DE

19. best supportive care

20. palliative therapy.DE

21. supportive therapy.DE

22. or/17-21

23. 16 or 22

24.  5 and 23

25.  ANIMAL/ OR NONHUMAN/ OR ANIMAL EXPERIMENT/

26.  HUMAN/

27.  26 and 25

28.  25 not 27

29.  24 not 28

1.3 Search of Cochrane Central Register of Controlled Trials (CENTRAL)

1. small cell lung cancer

2. OAT CELL

3. SCLC

4. 1 or 2 or 3

5. CONTROLLED CLINICAL TRIAL

6. RANDOM ALLOCATION

7. DOUBLE-BLIND METHOD

8. SINGLE-BLIND METHOD

9.  or/5-8

10. (ANIMAL NOT HUMAN)

11. 9 not 10

12.  CLINICAL TRIAL

13. (clin$ adj25 trial$).ti,ab

14. (( singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)). ti,ab

15. PLACEBO

16. placebo$.ti,ab

17. random$.ti,ab

18. RESEARCH DESIGN

19. or/12-18

20. 19 not 10

21. 11 or 20

22. 4 and 21

What's new

DateEventDescription
22 November 2013New citation required but conclusions have not changedThe review has been updated but conclusions have not changed.
17 October 2013New search has been performed

A new search was run and a new trial was identified in second-line chemotherapy at relapse, but conclusions did not change.

Some co-authors have changed.

History

Protocol first published: Issue 1, 2000
Review first published: Issue 4, 2003

DateEventDescription
5 August 2009New citation required but conclusions have not changedContact author changed as well as some co-authors.
3 August 2009New search has been performedFirst-line chemotherapy versus best supportive care or placebo is an update from the 2003 review. Second-line chemotherapy versus best supportive care in patients at relapse or progression has been added as a new objective. Searches were run, two new studies were identified but conclusions did not change.

Contributions of authors

Marta Pelayo, Virginie Westeel, Marcela Cortés-Jofré undertook all the parts of this update, with the help of Xavier Bonfill who reviewed the update and gave the necessary feedback.

Declarations of interest

None known

Sources of support

Internal sources

  • Iberoamerican Cochrane Centre, Spain.

External sources

  • Instituto Carlos III (Contract no. 10035) Ministry of Health, Spain.

    In the initial review of 2003

Differences between protocol and review

None

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Ciuleanu 2010

MethodsRandomised controlled trial
Participants

Randomised 401 patients, extensive disease refractory or in progression within 6 months of first-line chemotherapy

Evaluable 321, 84% male, median age 85 years, 88% Caucasian, 85% performance status 0-1, 15% performance status 2

Interventions

Arm A (n=133): BSC

Arm B (n=268): picoplatin IV 150 mg/m2 every 3 weeks median cycles 3 (1-15)

Outcomes

Median survival

Median survival in two subgroups (patients who did not received chemotherapy post-study, patients who were refractory or relapsed within 45 days of first-line chemotherapy)

Progression-free survival

Time to progression

Toxicity

Follow-up until death

Notes

Study published in abstract with complementary data on the Pharmaceutical website, without attrition data, tumour response, randomisation method. Sponsor: Poniard Pharmaceuticals

Uncertain/high risk of bias

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThere is no information data. Mentioned as multi-centre and probably randomisation made centrally. Patients randomly assigned 2:1
Allocation concealment (selection bias)Unclear riskThere is no information provided
Blinding of participants and personnel (performance bias)
Survival
Low riskNo information provided. Probably unblinded, but to determine survival it is probably not biased
Blinding of participants and personnel (performance bias)
Toxicity, tumour response, quality of life
Unclear riskNo information provided. Probably unblinded
Blinding of outcome assessment (detection bias)
Survival
Low riskNo information provided. Probably unblinded. Outcome measure is not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
Tumour response, toxicity, quality of life
Unclear riskNo information provided. Probably unblinded
Incomplete outcome data (attrition bias)
Tumour response, toxicity, quality of life
Unclear riskNo information provided in tumour response and lost to follow-up
Incomplete outcome data (attrition bias)
Survival
High riskNo Information provided
Selective reporting (reporting bias)High riskAuthors did not present all the information considered relevant
Other biasHigh riskNo information of baseline patient characteristics per group. Pharmaceutical sponsorship

Kokron 1977b

MethodsRandomised controlled trial
Participants

Randomised 34 patients with extensive disease

Evaluable 33, (only male), 70 years or less (median age 64.5, 66, and 63 years), no prior therapy, no brain metastases, no relevant secondary diseases, good general condition (able to work)

Interventions Arm A (n=17): supportive care.
Arm B (n=17): Ifosfamide: 1500 iv day 1, followed by 1000 iv days 2 to 3, repeated on week 2 and 3, with initial dose of 1000 iv day 1. Repetition after 6 weeks. Two cycles
Outcomes

Mean survival time
Tumour response (complete and partial response)
Toxicity

Follow-up: until death (no data)

Notes

Patients with extensive disease form a separate group

Bias with high probability

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe randomisation was conducted from the Institute of Cancer Research
Allocation concealment (selection bias)Unclear riskNo information provided
Blinding of participants and personnel (performance bias)
Survival
Low riskNo information provided. Probably unblinded, but to determine survival it is probably not biased
Blinding of participants and personnel (performance bias)
Toxicity, tumour response, quality of life
Unclear riskNo information provided. Probably unblinded
Blinding of outcome assessment (detection bias)
Survival
Low riskNo information provided. Probably unblinded. Outcome measure is not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
Tumour response, toxicity, quality of life
Unclear riskNo information provided. Probably unblinded
Incomplete outcome data (attrition bias)
Tumour response, toxicity, quality of life
Unclear riskNo information provided of patients lost to follow-up or toxic deaths
Incomplete outcome data (attrition bias)
Survival
Unclear riskNo Information provided
Selective reporting (reporting bias)Unclear riskAuthors did not present results on all important outcomes
Other biasUnclear riskNo information of baseline patient characteristics per group. Statistical analysis inadequate

Kokron 1982

MethodsRandomised controlled trial
ParticipantsRandomised 54 patients, 53 evaluable (32 extensive)
Extensive: Randomised 41, evaluable 32, (only male), 70 years or less (median age 62, 60.5, and 64 years), no prior therapy, no brain metastases, no relevant secondary diseases, good general condition (able to work)
Interventions Arm A (n=8): no active anticancer treatment, Ringer solution 1000 ml + 1 g vitamin C iv 3 times wk, 6 weeks
Arm B (n=12): Ifosfamide: 1500 iv day 1, followed by 1000 iv days 2 to 3, repeated on week 2 and 3, with initial dose of 1000 iv day 1
Repetition after 6 weeks
Arm C (n=12): Ifosfamide + CCNU (cyclohexylnitrosourea), Ifosfamide as in Arm B + CCNU 60 mg/m2 day 1
Repetition after 6 weeks
Outcomes

Mean survival time
Tumour response (complete and partial response)
Toxicity

Follow-up: until death (no data)

Notes

Patients with extensive disease form a subgroup

Bias with high probability

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe randomisation was conducted from the Institute of Cancer Research
Allocation concealment (selection bias)Unclear riskNo information provided
Blinding of participants and personnel (performance bias)
Survival
Low riskNo information provided. Probably unblinded, but to determine survival it is probably not biased
Blinding of participants and personnel (performance bias)
Toxicity, tumour response, quality of life
Unclear riskNo information provided. Probably partially blinded because a placebo infusion of Ringer's solution and Vitamin C was given
Blinding of outcome assessment (detection bias)
Survival
Low riskNo information provided. Probably unblinded. Outcome measure is not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
Tumour response, toxicity, quality of life
Unclear riskNo information provided. Probably unblinded
Incomplete outcome data (attrition bias)
Tumour response, toxicity, quality of life
Unclear riskNo information provided of patients lost to follow-up, toxic deaths. Tumour response is reported for extensive and limited disease combined
Incomplete outcome data (attrition bias)
Survival
Unclear riskNo Information provided
Selective reporting (reporting bias)Unclear riskAuthors did not present results in all important outcomes
Other biasUnclear riskNo information of baseline patient characteristics per group. Statistical analysis inadequate

O'Brien 2006

MethodsRandomised controlled trial
Participants

Randomised 141 patients at relapse (38 female)

Evaluable 141, > 18 years (median age = 58.6, 59.6 years), previous chemotherapy, no previous topotecan, no symptomatic brain metastases, no severe co-morbidities, performance status 0-2

Interventions

Arm A (n = 70): BSC

Arm B (n = 71): topotecan in capsules containing the equivalent to 0.25 mg, dosed 2.3 mg/m2 on days 1-5 every 21 days, at least 4 cycles + BSC                 

Outcomes

Median survival time
Tumour response (complete and partial response)
Toxicity

Quality of life

Follow-up: until death  (within 30 days of random assignment: 9 patients (13%) in arm A, 10 (7%)  in arm B

Attrition data: 33 in arm A, 21 in arm B

Notes

Compliance with topotecan was documented by returned capsule count (99% of patients took ≥ 90% of capsules)

Sponsor: GlaxoSmithKline Pharmaceuticals

Low risk of bias

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskPatients were randomly assigned 1:1
Allocation concealment (selection bias)Low riskCentralised automated registration
Blinding of participants and personnel (performance bias)
Survival
Low riskNo information provided. Probably unblinded, but to determine survival it is probably not biased
Blinding of participants and personnel (performance bias)
Toxicity, tumour response, quality of life
Unclear riskNo information provided. Probably unblinded
Blinding of outcome assessment (detection bias)
Survival
Low riskNo information provided. Probably unblinded. Outcome measure is not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
Tumour response, toxicity, quality of life
Unclear riskNo information provided. Probably unblinded
Incomplete outcome data (attrition bias)
Tumour response, toxicity, quality of life
Low riskPatients did not complete the study due to toxicity 9BSC, 13 topotecan, other 13 BSC
Incomplete outcome data (attrition bias)
Survival
Low riskinformation provided
Selective reporting (reporting bias)Low riskAuthors presented results on all outcomes measures that were pre-specified as relevant
Other biasLow riskThe study appears to be free of other bias. Pharmaceutical sponsorship

O'Brien 2007a

MethodsReport the results of quality of life of the O'Brien 2006 study
ParticipantsSame as O'Brien 2006
InterventionsSame as O'Brien 2006
OutcomesQuality of life
NotesIt is the same study as O'Brien 2006
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskSame as O'Brien 2006
Allocation concealment (selection bias)Unclear riskSame as O'Brien 2006
Blinding of participants and personnel (performance bias)
Survival
Low riskSame as O'Brien 2006
Blinding of participants and personnel (performance bias)
Toxicity, tumour response, quality of life
Unclear riskSame as O'Brien 2006
Blinding of outcome assessment (detection bias)
Survival
Low riskSame as O'Brien 2006
Blinding of outcome assessment (detection bias)
Tumour response, toxicity, quality of life
Unclear riskSame as O'Brien 2006
Incomplete outcome data (attrition bias)
Tumour response, toxicity, quality of life
Low riskSame as O'Brien 2006
Incomplete outcome data (attrition bias)
Survival
Low riskSame as O'Brien 2006
Selective reporting (reporting bias)Low riskSame as O'Brien 2006
Other biasLow riskSame as O'Brien 2006

Spiro 1989

MethodsRandomised controlled trial
ParticipantsIncluded 616 participants, 414 extensive. Randomised 390 patients on disease progression, (122 female). Evaluable 170 active treatment; no data in symptomatic; 75 years or less; (median age 62.5, 61.5 years); previous first-line chemotherapy, no severe co-morbidities, any performance status 
Interventions

Arm A (n= 198): symptomatic therapy  

Arm B (n= 192): methotrexate 50 mg/m2 iv plus doxorubicin 50 mg/m2 iv, repeated every 3 weeks up to a maximum of 9 courses                              

Outcomes

Median survival time
Tumour response (complete and partial response)
Toxicity

Follow-up: until death (no data)

Attrition data: 124 in arm B, no data in arm A 

NotesHigh risk of bias
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo information provided
Allocation concealment (selection bias)Unclear riskNo information provided
Blinding of participants and personnel (performance bias)
Survival
Low riskNo information provided. Probably unblinded, but to determine survival it is probably not biased
Blinding of participants and personnel (performance bias)
Toxicity, tumour response, quality of life
Unclear riskNo information provided. Probably unblinded
Blinding of outcome assessment (detection bias)
Survival
Low riskNo information provided. Probably unblinded. Outcome measure is not likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias)
Tumour response, toxicity, quality of life
Unclear riskNo information provided. Probably unblinded
Incomplete outcome data (attrition bias)
Tumour response, toxicity, quality of life
High riskPatients lost to follow-up not provided. Relapse chemotherapy was not administered to 124 patients, 54 (37.5%) assigned to four cycles of first-line chemotherapy and 70 (47%) assigned to eight of first-line chemotherapy. No toxicity was reported
Incomplete outcome data (attrition bias)
Survival
Unclear riskNo information provided
Selective reporting (reporting bias)Unclear riskAuthors did not present results on all outcomes measures considered relevant
Other biasLow riskNo information of baseline patient characteristics per group at treatment with second-line chemotherapy

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
AlIan 1984Phase II study of vinorelbine, no control group
Anderson 1993Ifosfamide-adryamicin in bolus plus oral etoposide versus continuous pump infusion of ifosfamide-adriamycin plus oral etoposide
Ardizzoni 2002Standard cyclophosphamide adriamycin etoposide versus intensified cyclophosphamide adriamycin etoposide
Beith 1996Induction chemoradiotherapy after randomisation to vincristine, doxorubicin,cyclophosphamide versus observation, in responding patients
Bleehen 1989Induction etoposide, cyclophosphamide, methotrexate, vincristine, after randomisation to further chemotherapy versus observation in partial or complete response
Bleehen 1993Etoposide, cyclophosphamide, methotrexate, vincristine versus etoposide, ifosfamide
Bleehen 1996Etoposide, cyclophosphamide, methotrexate, vincristine versus etoposide, vincristine
Candido 2002Phase II study of ifosfamide, epirubicin, carboplatin, no control group
Chen 2007Report the same quality of life results as O'Brien 2006, study included in the review
Ciuleanu 2002Phase II study of teniposide, etoposide, no control group
CuIlen 1986Induction therapy with vincristine, adriamycin, cyclophosphamide after randomisation to further courses or no more chemotherapy. Patients received as second-line treatment the same chemotherapy regimen of first-line treatment
EarI 1991Planned chemotherapy with cyclophosphamide, vincristine, etoposide versus chemotherapy given for tumour-related symptoms
Eridani 1978Adryamicin, cyclophosphamide versus bleomycin, adriamycin, cyclophosphamide, oncovicin combination versus cyclophosphamide, vincristine, cisplatin
Ettinger 1990Randomisation to cyclophosphamide, doxorubicin, vincristine vs the same regimen alternating with hexamethylmelamine, etoposide, methotrexate. Incomplete response randomisation to maintenance chemotherapy versus observation
Garassino 2011Retrospective report of outcomes in SCLC patients following various regimens of second-line chemotherapy
Giaccone 1993Induction cyclophosphamide, doxorubicin, etoposide. Incomplete or partial response randomisation to maintenance chemotherapy versus observation
Green 1969Cyclophosphamide vs placebo or nitrogen mustard versus placebo in previously treated patients with radiotherapy, surgical resection or previous chemotherapy not clarified
GrideIIi 2002Phase II study of carboplatin, vinorelbine, no control group
Hainsworth 2002Phase II study of paclitaxel, carboplatin, teniposide, etoposide, no control group
Hanna 2002Induction etoposide, ifosfamide cisplatin, and randomisation to oral etoposide or no further therapy in partial, complete or stable response
James 1996Cisplatin, etoposide, adriamycin, vincristine versus the same drugs at half the dose but twice the frequency
Joss 1995Carboplatin, teniposide versus cisplatin, adriamycin, etoposide alternating with cyclophosphamide, methotrexate, vincristine, carmustine
Kokron 1974aPreliminary results of the study of Kokron 1977b. Same as Kokron 1974b
Kokron 1974bPreliminary results of the study of Kokron 1977b
Kokron 1977aSame as Kokron 1977b
Korfel 2002Phase II study of teniposide as second-line chemotherapy in relapsed SCLC
Lebeau 1992Lomustine, cyclophosphamide, doxorubicin, etoposide plus aspirin versus same chemotherapy regimen. Incomplete response randomisation to same chemotherapy versus observation. Patients received more cycles of first-line treatment versus fewer cycles
Lyss 2002Phase II study of cisplatin, teniposide versus cisplatin, paclitaxel vs paclitaxel, teniposide versus paclitaxel teniposide at different doses
Mattson 1992Cyclophosphamide, vincristine, etoposide. Incomplete or partial response randomisation to interferon alfa versus cyclophosphamide, doxorubicin, cisplatin versus observation
Mattson 1997Cyclophosphamide, vincristine, etoposide. Incomplete or partial response randomisation to interferon alfa vs cyclophosphamide, doxorubicin, cisplatin versus observation
Maurer 1980Cyclophosphamide, methotrexate, vincristine pus cerebral irradiation vs cyclophosphamide. Incomplete response randomisation to further chemotherapy or observation. Patients received as second therapy more cycles of first-line chemotherapy versus fewer cycles
Mok 2002Phase I-II study of sequential administration of teniposide and oral etoposide
Naka 2002Phase II study of irinotecan and carboplatin for relapsed SCLC
Noda 2002Irinotecan, cisplatin versus etoposide cisplatin
Reck 1998Phase II study of bleomycin, no control group
Rosenthal 1991Phase II study of epirubicin in recurrent SCLC, no control group
Schiller 2001Cisplatin, etoposide. No progression randomisation to topotecan versus observation
Sculier 1996Ifosfamide, etoposide, anthracycline. Incomplete or partial responses randomisation to maintenance chemotherapy with etoposide, vindesine versus no further chemotherapy
Smit 1989Phase II study of etoposide in SCLC patients older than 70 years, no control group
Souhami 1997Oral etoposide versus cisplatin, etoposide intravenous
Sundstrom 2002Cisplatin, etoposide versus cyclophosphamide, etoposide, vincristine
Tada 2002Phase II study of carboplatin etoposide in SCLC patients older than 76 years, no control group
Thatcher 1996Oral etoposide versus etoposide, vincristine versus cyclophosphamide, adriamycin, vincristine
Thomas 2002Phase II study of CI-980, no control group
Tjan-Heíjnen 2001Cyclophosphamide, adriamycin, vincristine versus intensified same chemotherapy
Vanseenkiste 2001Phase II study of gemcitabine, etoposide, no control group
von Pawel 1990Phase II study of etoposide, vincristine, no control group
Zhang 2002Phase II study of teniposide, cisplatin, no control group

Ancillary