Sha Liu, and Guoli Zhang contributed equally to this work.
Prognostic factors and survival of patients with small cell lung cancer in a northeastern Chinese population
Article first published online: 22 APR 2013
© 2012 Tianjin Lung Cancer Institute and Wiley Publishing Asia Pty Ltd
Volume 4, Issue 2, pages 143–152, May 2013
How to Cite
Liu, S., Zhang, G., Li, C., Chen, X., Wang, S., Wang, M. and Cai, L. (2013), Prognostic factors and survival of patients with small cell lung cancer in a northeastern Chinese population. Thoracic Cancer, 4: 143–152. doi: 10.1111/j.1759-7714.2012.00149.x
- Issue published online: 22 APR 2013
- Article first published online: 22 APR 2013
- Accepted manuscript online: 21 MAY 2012 06:44AM EST
- Received: 24 February 2012; accepted 16 May 2012.
- lung cancer survival;
- lung cancer treatment;
- prognostic factor;
- small cell lung cancer
Background: This study presents the characteristics and treatment of small cell lung cancer (SCLC) and an analysis of the factors that impact survival in northeastern Chinese populations, among both smokers and non-smokers.
Methods: A retrospective review was performed using 485 Chinese patients diagnosed with pathologically confirmed SCLC diagnoses between January 2001 and December 2007. Data on patient characteristics, treatment patterns, and outcome information was collected systematically. Univariate analysis and the Cox multivariate regression model were used to evaluate prognostic factors.
Results: Median survival time was 16 months in all patients, 31 months in limited stage (LS) patients, and 10 months in extensive-stage (ES) patients. Never-smoking patients (P= 0.0368) with good performance status (PS) (P= 0.0044) or with normal lactate dehydrogenase (LDH) levels (P < 0.0001), demonstrated superior survival rates. Multivariate analysis identified that cycles of chemotherapy, PS, LDH levels, recurrence or progression, and clinical stage were each independent prognostic factors applicable to all patients. In LS-SCLC, cycles of chemotherapy were the only prognostic indicator; however, cycles of chemotherapy, LDH levels, and recurrence or progression, were all significant factors in ES-SCLC.
Conclusion: Cycles of chemotherapy, PS, LDH levels, recurrence or progression, and clinical stage were proved to be independent prognostic factors for SCLC with variant value based on the SCLC tumor stage.
Lung cancer remains a leading cause of cancer death in China, where it is most prominent in the industrialized northeast region, which is characterized by poor air quality and an increased prevalence of smoking habits. Small cell lung cancer (SCLC) contributes to about 13% of all new lung cancer diagnoses.1 SCLC is characterized by rapid doubling time, early dissemination, and high sensitivity to chemotherapy and radiotherapy, though surgery is only appropriate for few patients (2–5%).2 Two clinical stages of SCLC are generally recognized: limited-stage SCLC (LS-SCLC), and extensive-stage SCLC (ES-SCLC), according to the Veteran's Administration Lung Cancer Study Group.
Early combination chemotherapy treatments were pioneered in the 1970s, providing the basis for contemporary SCLC treatments, commonly administered as two-drug treatments of cisplatin/carboplatin plus etoposide in doses associated with at least moderate toxic effects, which often must take into account renal function and non-hematological toxic effects. In China, this method of treatment has been recognized as a superior chemotherapy regimen for both LS and ES-SCLC. Furthermore, combined modality treatments, wherein chemotherapy is combined with thoracic radiotherapy (TRT), are a current mainstay treatment for patients with LS-SCLC. Although the optimal period for initiation of radiotherapy remains a point of debate in modern research, increasing evidence has indicated the benefits of early TRT treatment. The National Cancer Institute of Canada (NCIC) reported a phase III trial in which TRT (40 Gy) treatment was initiated in either the second or last cycle of chemotherapy, providing strong evidence that early TRT treatment resulted in superior outcomes when compared to late TRT treatment.3 In patients exhibiting a complete response (CR) to treatment, prophylactic cranial irradiation (PCI) has been applied as a further method of inhibition of metastasis to tissues of the brain as well as other critical tissue regions. A recent report concluded that PCI was not only recommended to LS patients, but also to ES patients exhibiting positive initial responses to chemotherapy.4 In many cases, combination therapy can induce an initial response, showing initial effectiveness in limiting SCLC cell growth and proliferation.
Despite the initial effectiveness of modern chemotherapy and radiotherapy techniques, most patients with SCLC eventually exhibit regional relapse or distant metastases, accounting for the general poor prognosis and low long-term survival rate for the disease. In the absence of immediate treatment, the median survival rate for LS and ES-SCLC patients can be as low as 12 and 6 weeks, respectively.5 LS-SCLC patients treated with modern methods have a median survival of 13 to 34 months,6–8 while the median survival of ES-SCLC patients with proper therapy is only 7 to 11 months.7,9,10 Due to the relatively short survival period, clinical stage and performance status (PS) have been noted to be the most prominent prognostic factors.7,9 Other common prognostic factors include age, gender, smoking status, and lactate dehydrogenase (LDH) level, but the role of these factors in prognosis prediction in SCLC remains debated in contemporary research.
The characteristics and treatment patterns for SCLC in the northeast region of China, where the condition is notably prominent, are assessed in the current study. The association between these factors was analyzed in order to determine the factors that most prominently impact patient survival, providing a potentially useful tool in both research and clinical settings for patient assessment.
Patients and methods
Study cohort and data collection
This retrospective analysis was conducted using data from the Third Affiliated Hospital of the Harbin Medical University. A total of 485 patients with pathologically confirmed diagnosis of SCLC that had received hospital treatment between January 2001 and December 2007 were enrolled and actively followed. Of these 485 patients, 216 (44.54%) were diagnosed with LS-SCLC and 269 (55.46%) were diagnosed with ES-SCLC by disease stage. All patient diagnoses were histologically approved. The Ethics Committee of the hospital approved this study prior to its onset.
Age, gender, smoking status, pretreatment clinical stage, performance status (PS), date of original diagnosis, date of death or last observation, first-line treatment, tumor recurrence or progression, metastatic sites, hematological and biochemical test data was collected for each subject. PS ranged from zero to three on the Eastern Cooperative Oncology Group (ECOG) scale. Pretreatment clinical stage, metastases, and recurrence or progression were determined by physical examination, chest radiography, whole-body radionuclide bone scans, position emission tomography (PET) scans, and magnetic resonance imaging (MRI), coupled with computed tomography (CT) scans of the chest, brain, and abdomen.
According to the Veteran's Administration Lung Cancer Study Group criteria, patients were categorized as LS if the disease was confined to one thorax, including the ipsilateral mediastinal and supraclavicular nodes. Patients presenting the disease beyond these confines were categorized as ES. The time of disease progression was measured from original diagnosis to first evidence of disease progression. Overall survival was measured from the date of original diagnosis to death or last follow-up. Those patients who were lost or those who survived until the end of the follow-up period were regarded as censored.
Examined treatment types included chemotherapy, TRT, and chemoradiotherapy. Platinum-based chemotherapy was the standard therapy in the first-line setting of LS and ES-SCLC. Chemotherapy was categorized according to the regimen type: (1) cisplatin/carboplatin plus etoposide (EP or CE); (2) ifosfamide plus EP or CE (IEP or ICE); (3) epidoxorubicin plus EP or CE (EEP or ECE); and (4) all the other regimens or multi-combinations.
Chemoradiotherapy was defined as the concurrent use of chemotherapy and TRT. TRT was generally delivered over the course of four weeks, with cumulative doses ranging from 40 to 65 Gy administered in common does of two Gy each. Three-dimensional conformal radiotherapy was applied, occasionally combined with conventional radiotherapy. Early TRT treatment was defined as treatment initiating within the first or second cycles of chemotherapy.
The role of PCI was investigated wherein total doses ranged from 24 and 30 Gy delivered at three Gy per fraction per day. Complete remission (CR) was considered as the total clinical and radiologic resolution of the disease according to standard criteria. A reduction of at least 50% of all measurable lesions was regarded as a partial remission (PR). Other responses were cumulatively regarded as no response, such as stable disease (SD) or progressive disease (PD). Patients in CR were eligible for PCI treatment.
Statistical analysis was performed using SPSS 13.0 software. The Kaplan-Meier analysis was used to calculate survival curves and significant differences were determined bylog-rank testing. Univariate analysis was used to calculate the hazard ratios (HR) of each variable. Multivariate analyses were performed using Cox regression models to identify independent prognostic factors. A P-value of less than 0.05 (P < 0.05) was considered significant.
The two- and five- year survival rates observed in the 485 SCLC patients examined were 31.16% and 11.97%, respectively, with a median survival of 16 months. The clinical characteristics of these patients are listed in Table 1. Age distribution, sex difference, anatomical classification, and different chemotherapy regimens were analyzed, showing no statistical difference in outcome based on these factors. Notably, patients who had never smoked (n= 156, 32.16%) showed a trend toward better overall survival rates (P= 0.0368) (Table 1, Fig. 1). In addition, poor PS (two to three) was found to be a strong predictor of decreased survival (P= 0.0044) (Table 1, Fig. 2).
|Clinical characteristics||n (%)||Survival rate (%)||Median survival (months)||P-value|
|>300 × 109/L||128 (26.50)||23.22||0||14||0.1356|
|≤300 × 109/L||355 (73.50)||33.90||13.32||16|
|EP or CE||288 (62.47)||37.87||11.15||18||0.0525|
|IEP or ICE||47 (10.20)||24.60||9.11||15|
|EEP or ECE||32 (6.94)||30.85||20.56||16|
|Cycles of chemotherapy|
LS-SCLC and ES-SCLC stages
The median survival for patients with LS-SCLC and ES-SCLC was 31 and 10 months, respectively (Table 1), and the overall survival by stage is shown in Figure 3. The difference in the survival rate between the two groups was statistically significant (P < 0.0001), indicating that patients diagnosed with ES-SCLC generally exhibited poorer survival durations, often accompanied much earlier by severe complications, such as tissue metastases or pronounced cancer cell proliferation.
LDH and PLT serum levels
In the majority of patients (n= 480, 98.96%), LDH serum levels were available for analysis. Patients were grouped by LDH ≤ 249 and LDH > 249 U/L, the upper value considered in the normal range by the hospital. Patients with normal LDH levels showed better survival rates (P < 0.0001) (Table 1, Fig. 4). Furthermore, the frequency of elevated LDH levels in LS-SCLC and ES-SCLC was 29.44% (n= 63) and 51.88% (n= 138), respectively.
Platelet (PLT) serum levels were available in 483 (99.58%) patients. A PLT level higher than 300*109/L, referred to as thrombocytosis, was found in 26.50% (n= 128) of patients, based on a normal reference range of 100–300*109/L. The two- and five-year survival rates were 23.22% and 0.00%, respectively, in patients with elevated PLT levels, and 33.90% and 13.32%, respectively, in patients with normal PLT levels. This difference, however, was found to have no statistical significance (P= 0.1356). Anemia was also tested as a prognostic factor in this study, indicating no significant correlation (P= 0.075).
Metastases at diagnosis and follow-up
Distant metastasis at the time of diagnosis was observed in 143 (29.48%) patients. Among these patients, 32 (22.38%) patients exhibited multiple metastases. Brain (n= 35, 7.22%), liver (n= 34, 7.01%), bone (n= 62, 12.78%), and adrenal glands (n= 16, 3.30%) were the most common metastasis sites. Patients with brain or liver metastasis exhibited the shortest survival times, with a median of only nine and eight months, respectively. Patients with liver metastasis survived a maximum of two years. During follow-up, 153 (31.55%) patients exhibited developed metastasis with 55 (35.95%) of these patients exhibiting multiple metastases. The brain was found to be the most common metastatic site during follow-up, indicating a probability of brain metastases of 25.71% and 30.43% when survival time exceeded one year and two years, respectively.
Chemotherapy cycle number and TRT treatment
A total of 461 (95.05%) patients received chemotherapy. Of these patients, 244 (52.93%, n= 142 in LS, n= 102 in ES) patients received four to six cycles of chemotherapy, 177 (38.39%, n= 43 in LS, n= 134 in ES) patients received one to three cycles of chemotherapy, and 40 (8.68%, n= 25 in LS, n= 15 in ES) patients received more than six cycles of chemotherapy. Patients receiving low numbers of treatment cycles (one to three cycles), a treatment regime inferior in efficacy to extended chemotherapy cycles, generally experienced either financial constraints or severe adverse effects. The significant difference in survival rate corresponded to the number of cycles of chemotherapy in all patients, as well as both ES and LS patients (P < 0.0001) (Table 1, Fig. 5).
A total of 199 patients (n= 106 in LS, n= 93 in ES) were treated with sequential chemoradiotherapy, and 63 patients in the LS group and 55 patients in ES group were treated with early TRT. Results indicated that early TRT had no beneficial effect on survival in either LS (P= 0.1677) or ES (P= 0.8258) patients.
Only 10 (0.02%) patients were treated with PCI (n= 7 in LS and n= 3 in ES). Patients treated with PCI after initial therapy showed a better survival rate (P= 0.006), with no patients developing brain metastases by the two-year follow-up. Very few patients received PCI treatment, generally due to an unwillingness to undergo PCI treatment when no indications of brain metastasis presented. Patients treated with PCI were all complete responders.
Primary prognostic factors
Univariate analysis indicated that smoking habits, clinical stage, PS, recurrence or progression, metastasis at diagnosis, chemoradiotherapy, cycles of chemotherapy, and LDH levels, were the most important prognostic factors for SCLC. In patients with LS-SCLC, the primary prognostic factors for survival were chemoradiotherapy, cycles of chemotherapy, and LDH levels. Similarly, in patients with ES-SCLC, cycles of chemotherapy, recurrence or progression, LDH levels, and anemia, were most strongly associated with survival (Table 2).
|Patient characteristics||HR||95% CI||P-value|
|All : Smoking (Yes vs. No)||1.27||1.01,1.61||0.0423|
|Age at diagnosis (vs. <40)|
|Sex Women (vs. Men)||0.86||0.67,1.09||0.2179|
|Stage (Extensive vs. Limited)||4.00||3.16,5.05||<0.0001|
|PS (vs. 3)|
|Recurrent or progression (No vs. Yes)||0.54||0.43,0.67||<0.0001|
|Metastasis at diagnosis (Yes vs. No)||2.01||1.60,2.51||<0.0001|
|Treatment (vs. Chemoradiotherapy)|
|Cycles of chemotherapy (vs. 4–6)|
|Early TRT (no vs. yes)||0.84||0.66,1.07||0.1602|
|LDH (>249 U/L vs. ≤249 U/L)||1.86||1.50,2.30||<0.0001|
|PLT (>300 × 109/L vs. ≤300 × 109/L)||1.52||0.94,2.45||0.0844|
|Anemia (Yes vs. No)||1.19||0.94,1.51||0.1463|
|LS: Smoking (Yes vs. No)||1.36||0.91,2.02||0.1322|
|Age at diagnosis (vs. <40)|
|Sex Women (vs. Men)||0.81||0.53,1.25||0.3456|
|PS (vs. 3)|
|Recurrent or progression (No vs. Yes)||0.73||0.50,1.07||0.1046|
|Cycles of chemotherapy (vs. 4–6)|
|Treatment (vs. Chemoradiotherapy)|
|Early TRT (No vs. Yes)||1.06||1.09,2.37||0.7998|
|LDH (>249 U/L vs. ≤249 U/L)||1.61||0.93,2.12||0.0166|
|PLT (>300 X109/L vs. ≤300 × 109/L)||1.40||0.20,1.99||0.1087|
|Anemia (Yes vs. No)||0.63||0.69,1.61||0.4295|
|ES: Smoking (Yes vs. No)||1.12||0.84,1.49||0.4510|
|Age at diagnosis (vs. <40)|
|Sex Women (vs. Men)||0.75||0.55,0.99||0.0492|
|PS (vs. 3)|
|Metastasis at diagnosis (Yes vs. No)||0.97||0.74,1.25||0.7991|
|Cycles of chemotherapy (vs. 4–6)|
|Treatment (vs. Chemoradiotherapy)|
|Early TRT (No vs. Yes)||0.96||0.70,1.32||0.7887|
|Recurrent or progressive (No vs. Yes)||0.53||0.39,0.71||<0.0001|
|LDH (>249 U/L vs. ≤249 U/L)||1.40||1.08,1.83||0.0108|
|PLT (>300 × 109/L vs. ≤300 × 109/L)||1.04||0.78,1.39||0.7950|
|Anemia (yes vs. no)||2.04||1.20,3.48||0.0083|
The response effect demonstrated that CR, PR, SD, and PD rates were 22% (n= 48), 63% (n= 136), 11% (n= 24), and 3% (n= 7), respectively, in LS-SCLC patients. Only a single patient was lost to follow-up in this group. In ES-SCLC patients, the CR, PR, SD, PD rates were 5% (n= 13), 60% (n= 162), 16% (n= 42), and 13% (n= 36), respectively. In addition, 11 patients in this group died after one to three cycles of chemotherapy, and five additional patients were lost to follow-up.
In order to test for factors that might have an influence on overall survival, multivariate analyses using theCox-regression model were employed, including all statistically significant variables from univariate analysis. Cycles of chemotherapy, LDH levels, recurrence or progression, clinical stage, and PS, were observed to be independent prognostic factors for survival in all of the 485 patients. In LS-SCLC patients, the number of chemotherapy cycles was the only prognostic indicator. In ES-SCLC patients, anemia was excluded by multivariate analysis. Cycles of chemotherapy, LDH levels, and recurrence or progression, remained the most significant factors influencing survival rates (Table 3).
|Patients' characteristics||HR||95% CI||p-value|
|All: Cycles of chemotherapy (1–3 vs. 4–6)||3.18||2.48,4.09||<0.0001|
|LDH (>249 vs. ≤249)||1.51||1.21,1.89||0.0002|
|Recurrent or progression (No vs. Yes)||0.78||0.61,0.99||0.0467|
|Stage (Extent vs. Limited)||3.48||2.68,4.53||<0.001|
|PS (0 vs. 3)||0.71||0.57,0.90||0.0039|
|LS: Cycles of chemotherapy (1–3 vs. 4–6)||3.26||2.04,5.21||<0.0001|
|ES: Cycles of chemotherapy (1–3 vs. 4–6)||2.576||1.89,3.49||<0.0001|
|LDH (>249 vs. ≤249)||1.41||1.07,1.86||0.0148|
|Recurrent or progressive (no vs. yes)||0.702||0.51,0.96||0.0277|
The poor long-term survival of patients diagnosed with SCLC is indicative of the therapeutic challenge presented by this common cancer. The development of a reasonable prognostic system is required in order to optimize clinical treatment plans to increase the survival rate of SCLC patients. Because of the critical time factor involved in treatment initiation, a clear prognostic system is crucial to improving outcomes in these patients, in both Chinese and global populations.
Clinical stage and PS have been previously noted as the most prominent prognostic factors, though many other factors may have central roles that should also be considered.7,9 These factors have the potential to ensure accurate initial staging, providing the necessary differentiation between LS-SCLS and ES-SCLC treatment that is often not provided in many current clinical settings. Differences in median survival rates between LS-SCLC and ES-SCLC (31 vs. 10 months, P < 0.0001) were confirmed in this study, concurrent with previous research indicating the shorter survival times often observed in ES-SCLC patients as a result of the greater progression of their condition at diagnosis. Median survival was also shown to be significantly reduced in patients with poor PS (P= 0.0044); however, as PS is a subjective factor that can be affected by acute self-limited events, this may not be the best clinical indicator available.
Analysis of the Surveillance Epidemiology and End Results (SEER) database indicated that the percentage of SCLC patients among all lung cancer patients had decreased from 17.26% in 1986 to 12.95% in 2002 in the United States. This decrease is likely associated with a concomitant decrease in tobacco use and an increase in the prevalence of low-tar filter cigarettes.1 LS-SCLC patients may experience a greater chance of survival if they have never smoked, or if they quit smoking immediately after diagnosis.11 Videtic et al.12 also showed that LS-SCLC patients who continued cigarette smoking while receiving concurrent chemoradiotherapy exhibited significantly decreased survival rates. Brownson RC13 reported that more than 90% (97% in females and 91% in males) of SCLC patients were current or former smokers. Though the exact mechanisms for the negative effect exhibited by smoking history or smoking during treatment remains unclear, overwhelming evidence indicates that smoking increases the risk of development and progression of SCLC.
The current study indicated that life-long non-smokers (n= 156, 32.16%) had greater survival rates (P= 0.0368). Furthermore, univariate analysis showed a negative impact for patients with a smoking history (hazard ration [HR]: 1.27, P= 0.0423). Smoking, however, was not a statistically significant predictor in the multivariate analysis. The proportion (32.16%) of patients who had never smoked was higher in our data than in earlier published results. One limitation of the current study is the retrospective format, which may have introduced some error based on the number of included smoking patients versus never smoking patients. Further clinical studies may be devised to more closely examine the disease in specific populations of smoking or never smoking patients. Previously, the positive relationship between passive smoking and lung cancer has been confirmed by meta-analysis.14 In another meta-analysis, passive smoking was observed to be a significant risk factor for lung cancer in non-smoking Chinese individuals.15 Passive smoking and poor air quality due to pollution in northeast China were considered likely causative factors for the relative abundance of never-smoking patients presenting with SCLC in the population examined.
Some studies suggest that female gender is a favorable independent predictor,7,16,17 while others indicate no gender-based correlation.8 Wolf17 suggested that this effect exists only in female patients under 60 years of age. The current study suggests that female patients have greater survival times than males, however this result was not found to be statistically significant. Similarly, age has been debated as a predictor, and with several retrospective analyses confirming this finding.11,16 Increasing age was associated with decreasing PS and increased comorbidity.18 In Janssen-Heijnen's19 study of population-based data from SCLC patients aged ≥75 years, toxicity of chemotherapy was higher in patients with comorbidity and poor PS. Older patients were also more likely to deny treatment, which may have impacted the results of some studies, and current results support that age is not a prognostic factor when equivalent care is provided.7,10
LDH has been implicated in tumor progression, and various retrospective analyses have found that the pretreatment level of LDH in the serum is an independent prognostic predictor in both LS-SCLC and ES-SCLC,9,20 LS,21 or ES7. Based on elevated LDH in ES patients and the results of Cox multivariate analysis in the current study, LDH levels likely form an important independent predictor in ES patients. Previous studies have reported that thrombocytosis22,23 was associated with decreasing survival or impaired quality of life in lung cancer patients, possibly as a result of metastasis formation induced by a hyper-coagulability state through mechanisms such as stabilization of tumor emboli or promotion of tumor cell adhesion.24
In Ferrigno's study,25 an elevated PLT level was not an independent prognostic factor indicated via multivariate analysis. The current study indicated that patients with a PLT count ≥ 300 × 109 U/L exhibited the worst prognosis, though this finding was not statistically significant. Tumor anemia, leading to hypoxia in tumor tissues, common in cancer has been reported as a pre-treatment laboratory parameter closely related to patient survival, adversely affecting outcomes by reducing the efficiency of oxygen-dependent radiotherapy and chemotherapy.26–29 The current study indicates reduced survival times in patients with anemia, though further studies will be required to clarify the effect of thrombocytosis and anemia on the survival of SCLC patients.
As reported previously, the main sites of distant metastasis in our patients were brain, liver, bone, and adrenal glands.7,10,30 The frequency of brain metastasis was shown to increase from 7.22% at diagnosis to 30.43% at two years in the patients of the current study, a slightly lowered amount compared with previous reports that is potentially indicative of a unique variation in the Chinese population.31,32 Alternatively, this affect may be due to lower routine utilization of accurate imaging techniques in the asymptomatic patients in certain Chinese regions.
Micrometastatic movement of tumors to the brain is generally prevented by the blood-brain barrier, an eradicated structure virtually impervious to systemic chemotherapy. Meta-analyses have shown that PCI can further reduce the incidence of brain metastasis, increasing positive outcomes in SCLC patients.33,34 A recent review has shown that PCI significantly improves overall survival in both LS (from 15% to 20% at three years) and ES (from 13% to 27% at one year) patients.35 The current study confirms that patients treated with PCI after initial therapy showed increased survival rates (P= 0.006). While the current results confirm these findings, the proportion of patients treated with PCI was very low in the current study, as most patients denied PCI use, preventing a statistically significant analysis of PCI on survival rate. Previous studies have indicated that prolonged chemotherapy, beyond five cycles, does not offer a better chance of cure than short chemotherapy (five cycles). Increasing the number of cycles (up to six), however, was shown to be an indicator of increased survival time, with four to six cycles of chemotherapy being the optimal treatment observed in this study.36 Though the current study is limited in examination of whether longer chemotherapy cycles were the primary causative factor in improved outcomes, the low number of deaths during early treatment suggests that longer chemotherapy regimes play a role in better outcomes for surviving patients. Further studies will be required to assess the effectiveness of chemotherapy duration, intensity, and interval on early survival rates, as well as the long-term outcomes assessed by the current study. These results are concurrent with those of previous randomized trials conducted on SCLC patients.37–39
For LS-SCLC, combination chemotherapy and TRT is the accepted standard for treatment, reported to improve overall survival rates.2,40 The current univariate analysis indicates that the overall risk of death in the chemotherapy group compared with chemoradiotherapy group was 1.39 (P= 0.0209), though this effect was statistically insignificant in multivariate models. Previous meta-analysis has shown a small but significant improvement in two-year survival upon treatment with early TRT, with increasingly evident benefits using hyperfractionated TRT and platinum-based chemotherapy.41 The current study does not support the use of early TRT, likely due to the once-daily radiation treatment versus hyperfractionated TRT treatment. De Ruysscher42 indicated that duration from start to end of radiotherapy (SER) was the most important predictor of outcome, and shorter SER (<30 days) treatment was associated with improvements in survival rate. Although the supportive data for early TRT was not sufficient, it is widely accepted that concurrent chemoradiotherapy is superior to sequential therapy.
For ES-SCLC, chemotherapy is the standard treatment option.2 Although there is insufficient evidence to confirm the beneficial effect of TRT in ES-SCLC patients, it has been used as a consolidative therapy after chemotherapy in suitable patients. Oze43 analyzed 10 262 patients enrolled in 52 phase III trials between 1980 and 2006, concluding that survival rates had not improved significantly over the last three decades. The current study, similarly, demonstrates that concurrent chemoradiotherapy is not a prognostic factor for survival, highlighting the necessity of the development of new and novel treatment models for SCLC in order to extend patient survival duration and possibly provide a long-term treatment for this common form of cancer.
The characteristics and treatment patterns of SCLC in northeast China were evaluated, indicating that the number of cycles of chemotherapy, PS, LDH levels, recurrence or progression, and clinical stage, are each important independent prognostic factors in SCLC outcomes. In LS-SCLC, the most prominent prognostic factor was the number of cycles of chemotherapy, as indicated by improved outcomes in longer surviving patients. Similarly, in ES-SCLC, the most prominent prognostic factors were the number of cycles of chemotherapy, LDH levels, and recurrence or progression. Though the current retrospective study results suggest that the number of chemotherapy cycles can truly be considered as a true prognostic factor, further study will be required in order to assess the effects of early treatment survival rates, which may provide misleading correlations between outcome and cycle number. These results are a critical initial step towards providing a consistent means for patient evaluation throughout treatment and thus improvements in the design of clinical treatment plans for SCLC. Further studies will be required to develop new insights as to how this prognostic data may be applied to develop more effective therapeutic regimes at specific treatment intervals.
No authors declare any conflict of interest.
- 2NCCN (2009) Small Cell Lung Cancer. In: NCCN Clinical Practice Guidelines in Oncology V.2.2009. MS-1.
- 15[Meta-analysis of the relationship between passive smoking population in China and lung cancer]. Zhongguo Fei Ai Za Zhi 2010; 13: 617–23. [In Chinese., , et al.