Intervention Review

You have free access to this content

Interferon alpha for the adjuvant treatment of cutaneous melanoma

  1. Simone Mocellin1,*,
  2. Marko B Lens2,
  3. Sandro Pasquali1,
  4. Pierluigi Pilati1,
  5. Vanna Chiarion Sileni3

Editorial Group: Cochrane Skin Group

Published Online: 18 JUN 2013

Assessed as up-to-date: 22 AUG 2012

DOI: 10.1002/14651858.CD008955.pub2


How to Cite

Mocellin S, Lens MB, Pasquali S, Pilati P, Chiarion Sileni V. Interferon alpha for the adjuvant treatment of cutaneous melanoma. Cochrane Database of Systematic Reviews 2013, Issue 6. Art. No.: CD008955. DOI: 10.1002/14651858.CD008955.pub2.

Author Information

  1. 1

    University of Padova, Meta-Analysis Unit, Department of Surgery, Oncology and Gastroenterology, Padova, Veneto, Italy

  2. 2

    King's College london, Genetic Epidemiology Unit, London, UK

  3. 3

    Veneto Region Oncology Research Institute, Medical Oncology Unit 2, Padova, Italy

*Simone Mocellin, Meta-Analysis Unit, Department of Surgery, Oncology and Gastroenterology, University of Padova, Via Giustiniani 2, Padova, Veneto, 35128, Italy. simone.mocellin@unipd.it. mocellins@hotmail.com.

Publication History

  1. Publication Status: New
  2. Published Online: 18 JUN 2013

SEARCH

 
Characteristics of included studies [ordered by study ID]
Agarwala 2011

Methods553 centres enrolled participants

Accrual period: 1998 to 2010

Design: phase III randomised controlled trial (RCT)


ParticipantsNumber of participants: 1150

Observation N = 569 (group A); interferon alpha 2b N = 581 (group B)

Participants randomised: 1150

Participants evaluable: 1150

Inclusion criteria of the trial

  • participants with histologically proven melanoma > 1.5/2 mm thick or melanoma (> T3 tumour according to 6th and 7th edition of the AJCC TNM staging system)
  • participants with negative lymph node or with regional lymph metastasis detected with sentinel node biopsy (AJCC TNM stages N1a and N2a)


Exclusion criteria of the trial

  • evidence of distant metastasis
  • a history of prior malignant disease (except basal cell carcinoma of the skin or surgically treated early carcinoma of the cervix)
  • poor performance status
  • vital organ dysfunction


Interventions
  • Group A: observation
  • Group B: high-dose interferon alpha 2b (20 MU/m²) intravenously (i.v.) over 20 minutes daily for 5 consecutive days for 4 weeks


Outcomes
  1. Disease-free survival (DFS): No difference between arms was observed
  2. Overall survival (OS): No difference between arms was observed


Subgroup analysis: not performed


NotesMedian follow-up: not known

Participants with lymph node metastasis: histopathological status of lymph node was known in 1035 participants (90%). Sentinel node biopsy and elective lymph node dissection were performed in 955 (83%) and 122 (11%), respectively. Lymph node metastasis were detected in 101 (20%) and 95 (18%) in the observation and interferon arms, respectively

Dose reduction/treatment discontinuation: no available data

Quality of life: According to the protocol, it was assessed before treatment at day 22, every 3 months for 2 years, and then every 6 months for 3 years (results not reported)


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskThere was insufficient information about the sequence generation process to permit judgment

Allocation concealment (selection bias)Unclear riskThere was insufficient information to permit judgment

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskThere was insufficient reporting of attrition/exclusions to permit judgment

Selective reporting (reporting bias)Unclear riskThere was insufficient reporting of attrition/exclusions to permit judgment

Other biasLow riskThere was insufficient information to assess whether an important risk of bias existed, but we judged this study to be free of other sources of bias

Cameron 2001

Methods9 centres enrolled participants

Accrual period: not reported

Design: phase III randomised controlled trial (RCT)


ParticipantsNumber of participants: 96

Interferon alpha 2b N = 46 (group A); observation N = 49 (group B)

Participants randomised: 96

Participants evaluable: 94

Inclusion criteria of the trial

  • participants with histologically proven melanoma at least 3 mm thick or melanoma participants undergoing radical lymph node dissection (RLND) for regional lymph metastasis detected clinically and confirmed pathologically


Exclusion criteria of the trial

  • clinically positive lymph node
  • evidence of distant metastasis
  • a history of prior malignant disease in the last 5 years (except basal cell carcinoma of the skin or in situ tumours)
  • poor performance status
  • vital organ dysfunction
  • previous or concomitant chemotherapy, immunotherapy, and radiotherapy


Interventions
  • Group A: interferon alpha 2a, 3 MU given subcutaneously (s.c.) 3 times per week for 6 months
  • Group B: observation


Outcomes
  1. Disease-free survival (DFS): No difference between arms was observed
  2. Overall survival (OS): No differences between arms was observed


Subgroup analysis: not performed


NotesMedian follow-up: 78 months

Participants with lymph node metastasis: not reported

Dose reduction/treatment discontinuation: 3 participants (7%) required dose reduction (2 for neutropenia and 1 for drug-related fever); treatment was stopped in 13 (28%) cases following disease recurrence

Quality of life: not reported


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: "Patients were randomized"

Comment: There was insufficient information about the sequence generation process to permit judgment

Allocation concealment (selection bias)Unclear riskThere was insufficient information to permit judgment

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes

Other biasLow riskThe study appeared to be free of other sources of bias

Cascinelli 2001

Methods23 centres belonging to the World Health Organization (WHO) Melanoma Programme enrolled participants

Accrual period: 1990 to 1993

Design: phase III RCT


ParticipantsNumber of participants: 444

Interferon alpha 2a N = 225 (group A); observation N = 219 (group C)

Randomised: 444

Inclusion criteria of the trial

  • participants with primary melanoma and regional lymph node metastasis
  • age between 18 and 70 years
  • absence of residual disease at the primary site or distant sites beyond the draining nodes


Exclusion criteria of the trial

  • any other malignant disease (apart from basal cell carcinoma)
  • absence of cardiac, metabolic, neurological, or vascular disease (contraindicating long-term use of interferon alpha)
  • previous non-surgical treatment for melanoma
  • previous exposure to interferon alpha 2a
  • concomitant use of corticosteroids or other investigational drugs


Interventions
  • Group A: 3 MU s.c. of interferon alpha 2a 3 times per week for 3 years
  • Group B: observation


Outcomes
  1. Disease-free survival: not significant
  2. Overall survival: not significant


Subgroup analyses: not reported


NotesMedian follow-up: 88

Lymph node dissection was performed following the guidelines produced by the education project of the WHO Melanoma Programme

Participants with lymph node metastasis: 100%

Dose reduction/treatment discontinuation: not reported

Quality of life: Interferon has no negative effect on quality of life or daily activities of treated participants; fatigue and anxiety were more frequent in the interferon alpha 2a group than in those who had surgery alone


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: "eligible patients were randomly assigned"

Comment: There was insufficient information about the sequence generation process to permit judgment

Allocation concealment (selection bias)Low riskThe trial used central allocation

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes

Other biasLow riskThe study appeared to be free of other sources of bias

Creagan 1995

MethodsThe North Central Cancer Treatment Group and the Mayo Clinic conducted this study

Accrual period: 1984 to 1990

Design: phase III RCT


ParticipantsNumber of participants: 266

High-dose interferon alpha 2b N = 132 (group A); observation N = 132 (group B)

Participants randomised: 264

Participants evaluable: 264

Inclusion criteria of the trial

  • primary melanoma > 1.69 mm thick with clinically negative lymph nodes or positive nodes after lymphadenectomy


Exclusion criteria of the trial

  • any malignant tumour within the previous 3 years, except superficial squamous or basal cell carcinoma and in situ carcinoma of the cervix
  • evidence of unresected distant or regional disease


Interventions
  • Group A: high-dose interferon alpha 2a 20 MU/m² × 3/week for 4 months i.m.
  • Group B: observation


Outcomes
  1. Disease-free survival: There was no significant disease-free survival benefit for participants treated with interferon
  2. Overall survival: There was no significant overall survival benefit for participants treated with interferon


Subgroup analysis: There was a significant disease-free survival benefit for participants with lymph node metastasis


NotesMedian follow-up: 73 months

Lymph node dissection: Lymphadenectomy was performed in case of clinically positive lymph node

Participants with lymph node metastasis: 62%

Dose reduction/treatment discontinuation: More than half of participants (no quantification has been provided) required dose modification

Quality of life: not reported


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: "Patients randomized"

Comment: There was insufficient information about the sequence generation process to permit judgment

Allocation concealment (selection bias)Unclear riskThere was insufficient information to permit judgment

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes

Other biasLow riskThe study appeared to be free of other sources of bias

Eggermont 2005

MethodsThe trial was performed on behalf of the European Organization for Research and Treatment of Cancer (EORTC). 85 institutions in 22 countries enrolled participants

Accrual period: 1996 to 2000

Design: phase III RCT


ParticipantsNumber of participants: 1418

13-month interferon alpha 2b N = 553 (group A); 25-month interferon alpha 2b N = 556 (group B); observation N = 279 (group C)

Participants randomised: 1388

Inclusion criteria of the trial

  • age 18 to 75 years
  • AJCC stage IIb-III


Exclusion criteria of the trial

  • participants with mucosal or ocular melanoma
  • participants previously treated with systemic drugs for melanoma
  • participants with other malignant diseases (other than basal cell carcinoma, in situ cervical cancer), autoimmune disease, uncontrolled infections, cardiopulmonary disease, or liver or renal disease
  • participants taking corticosteroids


Interventions
  • Group A: 4 weeks of 10 MU of interferon (5 days per week), followed by 10 MU 3 times per week for 1 year
  • Group B: 4 weeks of 10 MU of interferon (5 days per week), followed by 5 MU 3 times per week for 2 years
  • Group C: observation


Outcomes
  1. Distant metastasis-free interval (time form randomisation to appearance of distant metastasis): There were no differences between participants enrolled in the treatment arms
  2. Distant metastasis-free survival (time from randomisation to distant metastasis or death): There were no differences between participants enrolled in the treatment arms


Subgroup analysis: node-negative participants (AJCC stage IIb) treated with interferon alpha 2b for 25 months had a better distant metastasis-free interval and survival than participants enrolled in the observation arm


NotesMedian follow-up: 56 months

Lymph node dissection: Regional lymph node dissection had to contain more than 5 nodes (inguinal), more than 10 nodes (axillary), or more than 15 nodes (neck)

Participants with lymph node metastasis: 75%

Dose reduction/treatment discontinuation: 16% of participants treated with 13-months of interferon and 20% of participants treated with 25-months of interferon

Quality of life: not reported


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "We enrolled and randomly assigned..."

Quote: "randomisation was done with minimisation techniques"

Allocation concealment (selection bias)Low riskQuote: "randomisation was done centrally from the EORTC data centre"

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes, including those that were prespecified

Other biasLow riskThe study appeared to be free of other sources of bias

Eggermont 2008

MethodsThe trial was performed on behalf of the EORTC. 99 centres in 17 countries (mainly in Europe) enrolled participants

Accrual period: information not reported

Design: phase III RCT


ParticipantsNumber of participants: 1256

Pegylated interferon alpha 2b N = 627 (group A); observation N = 629 (group B)

Participants randomised: 1256

Inclusion criteria of the trial

  • age 18 to 70 years
  • pathological AJCC stage III melanoma (TxN1–2M0)
  • adequate surgical margins at wide excision of primary melanoma
  • complete regional lymphadenectomy must have occurred 70 days or less before randomisation
  • normal liver, renal, and bone marrow function


Exclusion criteria of the trial

  • ocular or mucous membrane melanoma
  • evidence of distant metastasis or in-transit metastasis
  • prior malignancy within the past 5 years (other than surgically resected non-melanoma skin cancer or cervical carcinoma in situ)
  • autoimmune disease
  • uncontrolled infections, cardiovascular disease, liver or renal disease
  • use of systemic corticosteroids
  • previous use of systemic therapy for melanoma


Interventions
  • Group A: pegylated interferon alpha 2b was administered at 6 μg/kg per week s.c. for 8 weeks (induction phase), followed by 3 μg/kg per week s.c. for an intended treatment duration of 5 years (maintenance phase)
  • Group B: observation


Outcomes
  1. Disease-free survival: There was a significant difference in disease-free survival between treatment arms (the 4 years survival were 45.6% in the interferon group versus 38.9% in the observation group). Distant metastasis disease-free survival was not significantly different in participants treated with interferon or followed by observation
  2. Overall survival: No significant difference was seen in overall survival between the 2 groups


Subgroup analyses: Pegylated interferon alpha seemed effective in prolonging disease-free survival for participants with microscopic lymph node involvement (i.e. participants with positive sentinel node biopsy)


NotesMedian follow-up: 46 months

Lymph node dissection: Complete regional lymphadenectomy must have occurred 70 days or less before randomisation

Participants with lymph node metastasis: 100%

Dose reduction/treatment discontinuation: The study protocol specified stepwise dose adjustments (6 μg/kg per week to 3, 2, and 1 μg/kg per week during the induction phase, and from 3 μg/kg per week to 2 and 1 μg/kg per week during the maintenance phase) to adjust for toxicity and to maintain an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1 for each participant. Treatment could be interrupted for surgery for local or regional recurrence of melanoma, then resumed after surgery

Quality of life: Quality of life analysis, which has been reported elsewhere (Bottomley 2009), showed that pegylated interferon worsened patient quality of life. In particular, there were important differences for 5 scales used to assess quality of life: 2 functioning scales (social and role functioning) and 3 symptom scales (appetite loss, fatigue, and dyspnea), with the pegylated interferon alpha 2b arm being most impaired


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "patients were randomly assigned"

Quote: "randomisation was done with minimisation techniques; the sequence was generated by computer"

Allocation concealment (selection bias)Low riskQuote: "randomisation was done centrally at the EORTC data centre"

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes, including those that were prespecified

Other biasLow riskThe study appeared to be free of other sources of bias

Garbe 2008

Methods42 centres belonging to the Dermatologic Cooperative Oncology Group (DeCOG) network in Germany and Switzerland enrolled participants

Accrual period: 1997 to 2001

Design: phase III RCT


ParticipantsNumber of participants: 444

Interferon alpha 2a N = 148 (group A); interferon alpha 2a and dacarbazine (DTIC) N = 148 (group B); observation N = 148 (group C)

Randomised: 444

Evaluable: 441

Inclusion criteria of the trial

  • primary cutaneous malignant melanoma and pathologically proven regional node metastases (either microscopic or macroscopic metastasis)
  • complete lymphadenectomy and absence of satellite
  • in-transit or distant metastases
  • age between 18 and 75 years
  • absence of any other malignant disease apart from basal cell carcinoma
  • absence of cardiac, liver, renal, neurological, or autoimmune diseases


Exclusion criteria of the trial

  • previous exposure to interferon alpha
  • concomitant use of corticosteroids
  • other investigational drug


Interventions
  • Group A: 3 MU s.c. of interferon alpha 2a 3 times per week for 2 years
  • Group B: 3 MU s.c. of interferon alpha 2a 3 times per week for 2 years, plus DTIC 850 mg/m² every 4 to 8 weeks for 2 years
  • Group C: observation


Outcomes
  1. Disease-free survival: There was a significant improvement for participants treated with interferon alone with respect to those who received interferon + DTIC or observation
  2. Overall survival: There was a significant improvement for participants treated with interferon alone with respect to those who received interferon + DTIC or observation


Subgroup analyses: not reported


NotesMedian follow-up: 47 months

Lymph node dissection: This was carried out before randomisation, according to the German guidelines on the treatment of melanoma

Participants with lymph node metastasis: 100%

Dose reduction/treatment discontinuation: 21 participants withdrew their informed consent

Quality of life: EORTC QLQ-C30 questionnaire scores at baseline and 6 months after randomisation were compared in 238 participants. Participants under adjuvant treatment had a better outcome for 'physical functioning' (group A versus C: P = 0.007), 'role functioning' (group A versus C: P = 0.008), and 'emotional functioning' (group A versus C: P = 0.048) in comparison to the participants treated with surgery only


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "patients were randomly assigned"

Quote: "a permuted block randomization list"

Allocation concealment (selection bias)Low riskThe trial used central allocation

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes

Other biasLow riskThe study appeared to be free of other sources of bias

Grob 1998

Methods31 centres in France enrolled participants

Accrual period: 1990 to 1994

Design: phase III RCT


ParticipantsNumber of participants: 499

Interferon alpha 2a N = 244 (group A); observation N = 243 (group B)

Participants randomised: 499

Participants evaluable: 487

Inclusion criteria of the trial

  • age 18 to 75 years
  • histologically proven melanoma > 1.5 mm thick treated by wide excision


Exclusion criteria of the trial

  • clinically detectable regional node metastases
  • visceral metastases
  • previous treatment for melanoma except surgery
  • history of other cancer except basal cell carcinoma
  • white blood cell count < 4×10⁹/L, haemoglobin < 11 g/dL
  • creatinine, bilirubin, aspartate aminotransferase, alanine aminotransferase, and alkaline phosphatase concentrations < 1 to 5 times that of the upper limit of normal range


Interventions
  • Group A: interferon alpha 2a 3 MU s.c. 3 times per week for 18 months
  • Group B: observation


Outcomes
  1. Disease-free survival: Participants treated with interferon alpha 2a had a significantly longer DFS. The estimated hazard ratio showed a substantial reduction of relapse risk in treated participants as compared to controls during the first 2 years of follow-up
  2. Overall survival: No significant difference was observed between treatment arms


Subgroup analyses: not reported


NotesMedian follow-up: 60 months

Lymph node dissection: not reported

Participants with lymph node metastasis: Lymph node metastases were detected in 68 (28%) and 78 participants in the interferon and observation groups, respectively

Dose reduction/treatment discontinuation: Treatment was discontinued in 89 participants (36%): 43 relapsed; 7 refused further treatment; 3 were lost to follow-up; and 1 stopped because of a wrong diagnosis. 35 participants withdrew because of 1 or several adverse events

Quality of life: not reported


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "Patients were randomly assigned"

Quote: "The random-allocation list was computer generated"

Allocation concealment (selection bias)Low riskThe trial used central allocation

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes

Other biasLow riskThe study appeared to be free of other sources of bias

Hancock 2004

MethodsParticipants were enrolled in the UK

Accrual period: 1995 to 2000

Design: phase III RCT


ParticipantsNumber of participants: 674

Interferon alpha 2a N = 338 (group A); observation N = 336 (group B)

Participants randomised: 674

Participants evaluable: 633

Inclusion criteria of the trial

  • good performance status
  • primary melanoma resection performed within 12 weeks before enrolment


Exclusion criteria of the trial

  • history of other malignant disease
  • pregnancy or lactation
  • biological therapy
  • systemic corticosteroids or other immunosuppressive therapy


Interventions
  • Group A: interferon alpha 2a 3 MU s.c. 3 times per week for 24 months
  • Group B: observation


Outcomes
  1. Disease-free survival: There was no significant difference between the interferon arm and the control arm
  2. Overall survival: There was no significant difference between the interferon arm and the control arm


Subgroup analysis: Younger participants (age < 50 years) treated with interferon had a better DFS than younger participants untreated


NotesMedian follow-up: 36 months

Lymph node dissection: No description of the adopted surgical protocol was reported

Participants with lymph node metastasis: 85 participants (13%) had lymph node metastasis at the time of primary melanoma diagnosis

Dose reduction/treatment discontinuation: There were 50 withdrawals (14.8%) from interferon therapy due to toxicity; 24 participants stopped therapy for reasons other than toxicity, recurrence, or death from unrelated causes; dose reductions for toxicity were recorded for 21 participants (5 subsequently withdrew from therapy); 15 participants (3 subsequently withdrew) had breaks in therapy due to toxicity

Quality of life: Interferon has significant effects on quality of life and symptomatology and is unlikely to be cost-effective (Dixon 2006)


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "Patients were randomly assigned"

Quote: "Random assignments were balanced by minimization"

Allocation concealment (selection bias)Low riskThe trial used central allocation

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes

Other biasLow riskThe study appears to be free of other sources of bias

Hansson 2011

MethodsParticipants were enrolled by 35 centres in Nordic countries: 2 in Denmark, 5 in Finland, 6 in Norway, and 22 in Sweden. randomisation was done centrally at the data management centre (Karolinska University Hospital, Stockholm, Sweden)

Design: phase III RCT


ParticipantsNumber of participants: 855

Observation N = 284 (group A); 1-year treatment with interferon alpha 2b N = 285 (group B); 2-year treatment with interferon alpha 2b N = 286 (group C)

Participants randomised: 855

Participants evaluable: 855

Inclusion criteria of the trial

  • histologically verified resected cutaneous melanoma
  • AJCC stage IIB–IIC (T4N0M0) or stage III (TxN1–3M0)
  • age 18 years or older
  • ECOG performance status 0 to 1
  • normal bone marrow function
  • adequate liver chemistry and renal function


Exclusion criteria of the trial

  • non-cutaneous melanoma
  • melanoma with unknown primary site
  • incompletely resected melanoma
  • second cancer diagnosis (except for basal cell and squamous cell skin cancer or in situ cervical carcinoma)
  • systemic corticosteroid medication
  • pregnancy


Interventions
  • Group A: observation
  • Group B: interferon alpha 2b 10 MU flat dose s.c. 5 days per week for 4 weeks (induction), followed by interferon alpha 2b 10 MU flat dose s.c. 3 days per week for 12 months (maintenance)
  • Group C: interferon alpha 2b 10 MU flat dose s.c. 5 days per week for 4 weeks (induction), followed by interferon alpha 2b 10 MU flat dose s.c. 3 days per week for 24 months (maintenance)


Outcomes
  1. Disease-free survival: Median DFS was 23.2 months in group A, 37.8 months in group B, and 28.6 months in group C; the DFS difference between participants treated with interferon and those receiving observation was statistically significant (P = 0.034)
  2. Overall survival: Median OS differed between the groups (56.1 months in group A, 72.1 months in group B, and 64.3 months in group C), but the difference was not statistically significant (P = 0.60)


Subgroup analysis: Participants without ulcerated primary tumours and treated with interferon had benefit in terms of both DFS and OS. Participants with AJCC stage III disease with clinically palpable lymph node metastases had improved DFS if treated with interferon (P = 0.015)


NotesMedian follow-up: 72.4 months

Lymph node dissection: Sentinel node biopsy (SNB) was not performed in all enrolled participants as it was not routinely practised in enrolling countries. Elective lymph node dissection was performed in the majority of participants, whereas few participants were managed with sentinel node biopsy and completion lymph node dissection

Participants with lymph node metastasis: 80.6%

Dose reduction/treatment discontinuation: Dose reductions were reported in 164 (28.7%) of the 571 participants randomly assigned to interferon alpha 2b therapy: 77 (27.0%) in group B and 87 (30.4%) in group C. Interferon alpha 2b therapy was interrupted or discontinued prematurely in 147 participants (25.7%): 75 (26.3%) in group B and 72 (25.2%) in group C

Quality of life: assessed with the EORTC QLQ-C30 questionnaire at 9 time points from before randomisation to up to 2 years, with focus on the 6-month and 2-year assessments. It was significantly negatively affected in almost all areas during treatment, but levels similar to those in the control group were restored after the end of treatment


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "Patients were randomly assigned"

Quote: "block randomisation, with block size of 15"; "The allocation sequence was computer-generated"

Allocation concealment (selection bias)Low riskThe trial used central allocation

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was available, and all of the study's prespecified (primary and secondary) outcomes that were of interest in the review were reported in the prespecified way

Other biasLow riskThe study appeared to be free of other sources of bias

Kirkwood 1996

MethodsThe study was conducted on behalf of the Eastern Cooperative Oncology Group (ECOG) and United States (US) intergroup adjuvant studies

Accrual period: 1984 to 1990

Design: phase III RCT


ParticipantsNumber of participants: 287

High-dose interferon alpha 2b N = 147 (group A); observation N = 140 (group B)

Participants randomised: 287

Participants evaluable: 287

Inclusion criteria of the trial

  • histologically proven AJCC stage IIB or stage III primary or recurrent regional nodal involvement from cutaneous melanoma without evidence of systemic metastatic disease
  • normal organ function
  • no significant medical or psychiatric comorbidity
  • ECOG performance status of 0 or 1


Exclusion criteria of the trial

  • history of adjuvant radiotherapy, chemotherapy, and immunotherapy


Interventions
  • Group A: high-dose interferon alpha 2b for 1 year (20 MU/m²/d i.v. 5 days/week for 4 weeks; 10 MU/m² s.c. 3 days/week for 48 weeks)
  • Group B: observation


Outcomes
  1. Disease-free survival: There was a significant disease-free survival benefit for participants treated with interferon by comparison with observed participants
  2. Overall survival: There was a significant overall survival benefit for participants treated with interferon by comparison with observed participants


Subgroup analysis: not performed


NotesMedian follow-up: 83 months

Lymph node dissection: not available

Participants with lymph node metastasis: 89%

Dose reduction/treatment discontinuation: A dose reduction was observed in 35% of participants; treatment discontinuation was necessary in 34% and 41% of participants during induction and maintenance phases, respectively

Quality of life: Participants who received interferon alpha had more quality-of-life-adjusted survival than the observation group, regardless of the relative valuations placed on toxicity and relapse (Cole 1996)


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "Patients were randomized by permuted blocks"

Allocation concealment (selection bias)Unclear riskThere was insufficient information to permit judgment

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes

Other biasLow riskThe study appeared to be free of other sources of bias

Kirkwood 2000

MethodsThe study was conducted on behalf of the Eastern Cooperative Oncology Group (ECOG) and United States (US) intergroup adjuvant studies

Accrual period: 1991 to 1995

Design: phase III RCT


ParticipantsNumber of participants: 642

High-dose interferon alpha 2b N = 215 (group A); low-dose interferon alpha 2b N = 215 (group B); observation N = 212 (group C)

Participants randomised: 642

Participants evaluable: 608

Inclusion criteria of the trial

  • histologically proven AJCC stage IIB or stage III primary or recurrent regional nodal involvement from cutaneous melanoma without evidence of systemic metastatic disease
  • normal organ function
  • no significant medical or psychiatric comorbidity
  • ECOG performance status of 0 or 1


Exclusion criteria of the trial

  • history of adjuvant radiotherapy, chemotherapy, and immunotherapy


Interventions
  • Group A: high-dose interferon alpha 2b for 1 year (20 MU/m²/d i.v. 5 days/week for 4 weeks; 10 MU/m² s.c. 3 days/week for 48 weeks)
  • Group B: low-dose interferon alpha 2b (3 MU/d 3 days/week)
  • Group C: observation


Outcomes
  1. Disease-free survival: 5-year DFS was 44% in group A, 40% in group B, and 35% in group C. High-dose interferon alpha 2b treatment is significantly superior to observation, while no statistically significant difference is observed between the latter and low-dose interferon alpha 2b
  2. Overall survival: No significant difference was observed between treatment arms and observation


Subgroup analysis: High-dose interferon alpha 2b was associated with a survival benefit in participants with 2 to 3 positive lymph nodes


NotesMedian follow-up: 52 months

Lymph node dissection: Participants with tumour depth > 4 mm and no clinical evidence of lymph node metastasis were not required to undergo lymphadenectomy

Participants with lymph node metastasis: 75%

Dose reduction/treatment discontinuation: 58% of participants treated with high-dose interferon alpha 2b required delay or dose reduction (44% due to toxicity). A significantly larger proportion of relapsed participants from the observation arm (31%) received an interferon-containing salvage regimen compared with only 15% of participants from the high-dose interferon alpha 2b arm

Quality of life: 77% of participants would experience a benefit in quality-of-life-adjusted-survival from interferon alpha, and 23% would experience a decrease in quality-of-life-adjusted-survival, although these effects did not reach statistical significance (Kilbridge 2002)


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskQuote: "Patients were randomized by permuted blocks"

Allocation concealment (selection bias)Unclear riskThere was insufficient information to permit judgment

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes

Other biasLow riskThe study appeared to be free of other sources of bias

Kirkwood 2001

MethodsThe study was conducted by Eastern Cooperative Oncology Group (ECOG) and United States (US) intergroup adjuvant studies

Accrual period: 1996 to 1999

Design: phase III RCT


ParticipantsNumber of participants: 880

GMK N = 389 (group A); interferon alpha 2b N = 385 (group B)

Participants randomised: 774

Participants evaluable: 774

Inclusion criteria of the trial

  • age > 18 years
  • histologically proven AJCC stage IIB/III primary cutaneous melanoma or clinically detected nodal metastasis arising from an unknown primary or a first clinically detectable nodal recurrence
  • without evidence of systemic metastases
  • primary melanoma > 4.0 mm thick with microscopic satellite lesions within 2 cm of the primary tumour
  • normal organ function
  • absence of significant medical or psychiatric comorbidity
  • an ECOG performance status of 0 or 1


Exclusion criteria of the trial

  • participants with gross subcutaneous invasion or grossly apparent satellite lesions
  • prior adjuvant radiotherapy, chemotherapy, or immunotherapy
  • contraindication to interferon usage
  • pregnant or lactating women


Interventions
  • Group A: 1 ml of GMK vaccine administered s.c. on days 1, 8, 15, and 22, then every 12 weeks (weeks 12 to 96)
  • Group B: high-dose interferon alpha 2b, 20 MU/m²/d i.v. 5 days/week x 4 weeks followed by 10 MU/m² s.c. 3 days/week x 48 weeks


Outcomes
  1. Disease-free survival: High-dose interferon alpha 2b demonstrated a significant DFS benefit with respect to GMK
  2. Overall survival: High-dose interferon alpha 2b demonstrated a significant OS benefit with respect to GMK


Subgroup analyses: Node-negative participants had a statistically significant DFS and OS benefit compared with participants treated with GMK


NotesMedian follow-up: 16 months

Lymph node dissection: It was performed in case of clinically positive lymph node or positive sentinel lymph node

Participants with lymph node metastasis: Lymph node involvement was documented clinically or pathologically in 77% of participants

Dose reduction/treatment discontinuation: 45 participants in the interferon alpha 2b arm (10%) discontinued treatment because of adverse events, while no participants experienced treatment discontinuation, dose reduction, or both, in the GMK arm

Quality of life: not reported


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: "patients were randomized"

There was insufficient information about the sequence generation process

Allocation concealment (selection bias)Low riskThe trial used central allocation

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes

Other biasLow riskThe study appeared to be free of other sources of bias

Kirkwood 2001a

MethodsThe study was conducted on behalf of the Eastern Cooperative Oncology Group (ECOG) and United States (US) intergroup adjuvant studies

Accrual period: -

Design: phase II RCT


ParticipantsNumber of participants: 107

Interferon alpha 2b and GMK vaccine starting on day 1 N = 36 (group A); GMK vaccine starting from day 1 and interferon alpha 2b starting from week 5 N = 36 (group B); GMK vaccine alone N = 35 (group C)

Participants randomised: 107

Participants evaluable: 107

Inclusion criteria of the trial

  • age > 18 years
  • free of disease after complete surgical resection for AJCC stage IIB, III, or IV melanoma (including participants with resectable distant metastatic disease, regionally advanced in-transit metastases, and extracapsular extension of nodal disease)
  • participants with AJCC stage IIB-III disease ineligible for E1694 because more than 56 days had elapsed since surgery
  • enter this study within 1 year after surgery
  • ECOG performance status < 2
  • normal white blood cell and platelet counts
  • AST (aspartate transaminase) and bilirubin ≤ 2 times the normal values


Exclusion criteria of the trial

  • history of any prior systemic anticancer therapy (including immunemodulators)
  • concomitant autoimmune or malignant disease
  • anti-inflammatory immunosuppressive or antihistaminic drugs (including corticosteroids)
  • previous splenectomy
  • history of heart disease higher than New York Heart Association (NYHA) class 2
  • organic brain syndrome, neuropathy, or active infection
  • history of severe allergic reaction to shellfish


Interventions
  • Group A: high-dose interferon alpha 2b, 20 MU/m²/d i.v. 5 days per week x 4 weeks, followed by 10 MU/m² s.c. 3 days per week x 48 weeks 1 mL plus GMK was administered s.c. on weeks 1, 2, 3, 4, 12, 24, and 36 (each dose contained 30 µg of GM2 and 100 µg of QS21). Both drugs were started on day 1
  • Group B: high-dose interferon alpha 2b, 20 MU/m²/d i.v. 5 days per week x 4 weeks, followed by 10 MU/m² s.c. 3 days per week x 48 weeks 1 mL plus GMK was administered s.c. on weeks 1, 2, 3, 4, 12, 24, and 36 (each dose contained 30 µg of GM2 and 100 µg of QS21). GMK vaccine was started on day 1, while interferon alpha 2b was started after 5 weeks
  • Group C: GMK was administered s.c. on weeks 1, 2, 3, 4, 12, 24, and 36 (each dose contained 30 µg of GM2 and 100 µg of QS21)


Outcomes
  1. Disease-free survival: GMK combined with interferon alpha 2b (arms A and B) was associated with longer survival compared to participants who received GMK alone
  2. Overall survival: No significant difference was observed between treatment arms (data suitable for meta-analysis were not reported)


Subgroup analysis: not performed


NotesMedian follow-up: 24 months

Lymph node dissection: not reported

Participants with lymph node metastasis: 89%

Dose reduction/treatment discontinuation: 17 participants (16%) discontinued treatment (12 because of toxicity, and 5 withdrew consent)


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: "Patients were assigned randomly"

There was insufficient information about the sequence generation process

Allocation concealment (selection bias)Low riskThe trial used central allocation

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes

Other biasLow riskThe study appeared to be free of other sources of bias

Kleeberg 2004

MethodsParticipants were randomised by 45 institutions in 13 countries within the frame of the EORTC 18871 3-arm trial (comparing observation versus interferon alpha 2b versus interferon gamma) and the DKG-80-1 4-arm trial (comparing observation versus interferon alpha versus interferon gamma versus Iscador-M)

Accrual period: 1988 to 1996

Design: phase III RCT


ParticipantsNumber of participants: 830

EORTC 18871 3-arm trial (N = 423): observation N = 142; interferon alpha 2b N = 139; interferon gamma N = 142

DKG-80-1 4-arm trial (N = 407): observation N = 102; interferon alpha 2b N = 101; interferon gamma N = 102; Iscador-M N = 102

In order to ascertain interferon's role in the adjuvant treatment of high-risk melanoma participants, the results of these 2 trials were pooled together: Overall, there were 244 participants in the observation arm, 240 in the interferon alpha 2b arm, and 244 in the interferon gamma arm

Participants randomised: 830

Participants evaluable: 793

Inclusion criteria of the trial

  • age 14 to 80
  • primary melanoma thick 3 mm or presence of lymph node metastasis regardless of primary tumour thickness
  • primary tumour resected with at least 2 cm surgical margin


Exclusion criteria of the trial

  • not reported


Interventions
  • Interferon alpha 2b: 1 MU s.c. every other day for 1 year
  • Interferon gamma: 0.2 mg s.c. every other day for 1 year
  • Iscador-M: Treatment was started at "dose level 0", and the dose was escalated from 0.01 to 1.0 mg/ml, every other day, over 2 weeks. After 3 days without treatment, injections were resumed for 14 doses (28 days) of 20 mg/ml followed by 7 days of no treatment


Outcomes
  1. Disease-free survival: There was no significant difference between interferon and observation arm
  2. Overall survival: There was no significant difference between interferon and observation arm


NotesMedian follow-up: not reported

Lymph node dissection: Clinically lymph node-positive participants received radical lymph node dissection, while clinically node-negative participants were treated with elective lymph node dissection or followed by observation based upon each institution's policy

Participants with lymph node metastasis: 58%

Dose reduction/treatment discontinuation: 11 participants (4.6%) treated with interferon alpha 2b, 19 (7.8%) treated with interferon gamma, and 5 (4.9%) in the Iscador-M arm due to toxicity

Quality of life: not reported


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: "Patients were randomized"

There was insufficient information about the sequence generation process

Allocation concealment (selection bias)Low riskThe trial used central allocation

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes

Other biasLow riskThe study appeared to be free of other sources of bias

McMasters 2008

MethodsThe Sunbelt Melanoma Trial involved 79 centres across North America

Accrual period: 1997 to 2003

Design: phase III RCT


ParticipantsNumber of participants: 774

Protocol A: Participants with only 1 pathologically positive sentinel lymph node (SLN) who underwent completion lymph node dissection (CLND) were randomised to observation (group A) versus high-dose interferon therapy (group B)

Protocol B: Histologically negative SLN were analysed by reverse transcription polymerase chain reaction (PCR); participants with PCR-positive SLN were randomised to 3 treatment arms: observation (group C), CLND (group D), or CLND plus interferon (group E)

Participants randomised: 774

Participants evaluable: 774

Inclusion criteria of the trial

  • age 18 to 70 years
  • cutaneous melanomas > = 1 mm thick


Exclusion criteria of the trial

  • evidence of regional or distant metastasis


InterventionsProtocol A (participants with 1 positive sentinel lymph node submitted to CLND)

  • Group A: observation
  • Group B: interferon alpha 2b 20 MU/m² i.v. per day, 5 days per week × 4 weeks followed by 10 MU/m² s.c. 3 times per week for 48 weeks


Protocol B (participants with histologically negative but PCR-positive SLN)

  • Group C: observation
  • Group D: completion lymph node dissection (CLND)
  • Group E: interferon alpha 2b 20 MU/m² i.v. per day, 5 days per week × 4 weeks (1997 to 1999). Since 1999, this schedule was followed by 10 MU/m² s.c. 3 times per week for 48 weeks


Outcomes
  1. Disease-free survival: There was no significant difference between interferon and observation arm in either protocols
  2. Overall survival: There was no significant difference between interferon and observation arm in either protocols


Subgroup analysis: A paper reporting a subgroup analysis from this trial investigated the association between primary tumour ulceration and regional lymph node status, showing that interferon treatment was associated with improved disease-free survival in participants with ulcerated primaries and lymph node metastasis (McMasters 2010)


NotesMedian follow-up: 64 months

Lymph node dissection: All participants underwent SLN biopsy; CLND was performed according to the above reported criteria

Participants with lymph node metastasis: 18%

Dose reduction/treatment discontinuation: not reported

Quality of life: not reported


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskThere was insufficient information about the sequence generation process

Allocation concealment (selection bias)Unclear riskThere was insufficient information to permit judgment

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskThere was insufficient reporting of attrition or exclusions to permit judgment

Selective reporting (reporting bias)Unclear riskThere was insufficient information to permit judgment

Other biasLow riskThere was insufficient information to assess whether an important risk of bias exists, but we judged this study to be free of other sources of bias

Pehamberger 1998

MethodsThe Austrian Melanoma Cooperative Group (7 Institutions) conducted the study

Accrual period: 1990 to 1994

Design: phase III RCT


ParticipantsNumber of participants: 311

Interferon alpha 2a N = 154 (group A); observation N = 157 (group B)

Participants randomised: 311

Participants evaluable: 293

Inclusion criteria of the trial

  • age 18 to 75 years
  • Breslow thickness > 1.5 mm with negative lymph node (AJCC TNM stage II participants)


Exclusion criteria of the trial

  • evidence of distant, lymph node metastasis, or both
  • secondary neoplasm


Interventions
  • Group A: interferon alpha 2a 3 MU x 7/week (3-week, route: s.c.) + 3 MU s.c. x 3/week (12 months, route: s.c.)
  • Group B: observation


Outcomes
  1. Disease-free survival: Interferon alpha 2a was found to provide a significant survival advantage
  2. Overall survival: not investigated


Subgroup analysis: No significant subgroup of enrolled participants had a benefit after interferon treatment


NotesMedian follow-up: 41 months

Lymph node dissection: Elective lymph node dissection was not performed

Participants with lymph node metastasis: 0%

Dose reduction/treatment discontinuation: A dose reduction was necessary in 8 treated participants and a discontinuation of treatment in 12

Quality of life: not reported


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskQuote: "patients were randomly assigned"

There was insufficient information about the sequence generation process

Allocation concealment (selection bias)Unclear riskThere was insufficient information to permit judgment

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data were balanced in numbers across intervention groups, with similar reasons for missing data across groups

Selective reporting (reporting bias)Low riskThe study protocol was not available, but it is clear that the published reports include all expected outcomes

Other biasLow riskThe study appeared to be free of other sources of bias

Rusciani 1997

MethodsThe study was conducted by the Catholic University and Sapienza University, Rome, Italy

Accrual period: not available

Design: phase III RCT


ParticipantsNumber of participants: 154

Interferon alpha 2b N = 84 (group A); observation N = 70 (group B)

Participants randomised: 154

Participants evaluable: 154

Inclusion criteria of the trial

  • participants with cutaneous melanoma and clinically negative lymph node


Exclusion criteria of the trial

  • not reported


Interventions
  • Group A: interferon alpha 2b 3 MU × 3/week (3-week, route: i.m.) for cycles of 6 months with 1-month interval between cycles for 3 years
  • Group B: observation


OutcomesThe crude incidence of recurrence was lower for participants treated with interferon alpha than participants enrolled in the control arm.

The analysis were not performed following a methodology suitable to investigate survival


NotesMedian follow-up: not reported

Lymph node dissection: not performed

Participants with lymph node metastasis: 0%

Dose reduction/treatment discontinuation: Treatment suspension or discontinuation was never required because of adverse events

Quality of life: not reported


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskQuote: "The authors have started a randomized clinical trial", "concomitant patients"

Comment: We contacted the authors of the study by telephone, and they confirmed the randomised design of the study

There was no information about the sequence generation process

Allocation concealment (selection bias)High riskThere was no information about allocation concealment

Blinding of participants and personnel (performance bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThere was no blinding, but we judged that lack of blinding was not likely to influence the outcome

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskThere was insufficient reporting of attrition/exclusions to permit judgement of 'low risk' or 'high risk' (e.g. number randomised not stated; no reasons for missing data provided)

Selective reporting (reporting bias)High riskOutcomes of interest in the review were reported incompletely, so we could not enter them in a meta-analysis. The study report fails to include results for disease-free survival, overall survival, or both, which would be expected to have been reported for such a study

Other biasLow riskThe study appeared to be free of other sources of bias

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Anaya 2008This was a non-randomised study

Chiarion-Sileni 2011As this trial compared 2 different schedules/dosages of adjuvant interferon alpha (i.e. intensified dose of interferon alpha versus standard high dose), it was not suitable for investigating whether or not interferon is superior to any comparator other than interferon for the adjuvant treatment of high-risk melanoma participants

Dillman 2003Participants enrolled in this study received interferon gamma and not interferon alpha, which is the drug under investigation in this review

Grob 2010As this trial compared 2 different schedules and types of interferon (36 months of pegylated interferon versus 18 months of low-dose interferon), it was not suitable for investigating whether or not interferon is superior to any comparator other than interferon for the adjuvant treatment of high-risk melanoma participants

Hauschild 2003This study randomised participants to low-dose interferon in a combination regimen with low-dose interleukin-2 or to observation. The presence of interleukin-2 (an active drug in metastatic melanoma) in the treatment schedule might affect interferon efficacy, which precluded inclusion of this trial

Hauschild 2009As this trial compared 2 different schedules/dosages of adjuvant interferon alpha (i.e. low-dose interferon alpha 2b with or without a modified high-dose interferon alpha 2b induction phase), it was not suitable for investigating whether or not interferon is superior to any comparator other than interferon for the adjuvant treatment of high-risk melanoma participants

Hauschild 2010As this trial compared 2 different schedules/dosages of adjuvant interferon alpha (i.e. low-dose interferon alpha 2a for 18 versus 60 months), it was not suitable for investigating whether or not interferon is superior to any comparator other than interferon for the adjuvant treatment of high-risk melanoma participants

Kerin 1995As this trial compared interferon combined with dacarbazine versus observation, it was not suitable for investigating whether or not interferon is superior to any comparator other than interferon for the adjuvant treatment of high-risk melanoma participants

Kim 2009This was a 2-arm randomised trial comparing biochemotherapy (including interferon) versus interferon in high-risk melanoma participants. As interferon was included in both treatment arms, this trial was not suitable for inclusion in this review

Kokoschka 1990The randomisation of the trial was unclear. Even though the authors mentioned that participants were randomly selected to receive interferon alpha therapy, there was no mention of a corresponding randomised control arm

Mao 2011This phase II randomised study enrolled participants with acral melanoma and compared different interferon schedules (i.e. 1 month versus 1 year high-dose interferon adjuvant treatment), so it did not meet the inclusion criteria of this review

Meyskens 1995Participants enrolled in this study received interferon gamma and not interferon alpha, which is the drug under investigation in this review

Mitchell 2007Participants of this study were randomly assigned to receive low-dose interferon alpha plus specific immunotherapy with allogeneic melanoma lysates or high-dose interferon alpha. As interferon was included in both treatment arms, this trial was not suitable for investigating whether or not interferon is superior to any comparator other than interferon for the adjuvant treatment of high-risk melanoma participants

Mohr 2008As this trial compared 2 different schedules/dosages of adjuvant interferon alpha (i.e. intensified dose of interferon alpha versus standard high dose), it was not suitable for investigating whether or not interferon is superior to any comparator other than interferon for the adjuvant treatment of high-risk melanoma participants

Pectasides 2009As this trial compared 2 different schedules of adjuvant interferon alpha (1 month versus 1 year), it was not suitable for investigating whether or not interferon is superior to any comparator other than interferon for the adjuvant treatment of high-risk melanoma participants

Richtig 2005Participants with AJCC stage II melanoma were randomly assigned to receive interferon alpha at a dosage of 3 megaunits 3 times each week for 2 years, plus isotretinoin at a dose of 20 mg for participants < or = 73 kg, 30 mg for participants greater than 73 kg, versus interferon alpha at the same dosage plus placebo. As interferon alpha was scheduled in both arms, it was not eligible for inclusion

Rusciani 2007This was a non-randomised study

 
Characteristics of ongoing studies [ordered by study ID]
ECOG E1609

Trial name or titleA Phase III Randomized Study of Adjuvant Ipilimumab Anti-CTLA4 Therapy Versus High-Dose Interferon Alpha-2b for Resected High-Risk Melanoma (clinicaltrial.gov identifier: NCT01274338)

MethodsThis is a phase III randomised study

Participants
  • High-risk stage III or stage IV melanoma that has been removed by surgery

Interventions
  • Experimental: Participants receive ipilimumab IV over 90 minutes on day 1. Treatment repeats every 21 days for a total of 4 courses in the absence of disease progression or unacceptable toxicity. Beginning on week 24, participants receive maintenance ipilimumab IV over 90 minutes on day 1. Treatment repeats every 90 days for a maximum of 4 courses in the absence of disease progression or unacceptable toxicity
  • Active comparator: Participants receive high-dose recombinant interferon alpha 2b IV on days 1 to 5, 8 to 12, 15 to 19, and 22 to 26 in the absence of disease progression or unacceptable toxicity. Participants then receive maintenance high-dose recombinant interferon alpha 2b subcutaneously on days 1, 3, and 5. Treatment repeats every week for 48 weeks in the absence of disease progression or unacceptable toxicity

OutcomesPrimary outcomes of the trial

  1. Recurrence-free survival
  2. Overall survival


Secondary outcomes of the trial

  1. Toxicity
  2. Quality of life

Starting dateMay 2011

Contact informationPrincipal Investigator: Ahmad Tarhini, Eastern Cooperative Oncology Group

NotesExpected primary completion date: May 2018

EORTC 18081

Trial name or titleAdjuvant Pegylated-Interferon-alpha2b (SylatronTM) for 2 Years Versus Observation in Patients With an Ulcerated Primary Cutaneous Melanoma With T(2-4)bN0M0: a Randomized Phase III Trial of the EORTC Melanoma Group (clinicaltrial.gov identifier: NCT01502696)

MethodsThis is a phase III randomised study

Participants
  • Patients with an ulcerated melanoma with Breslow > 1 mm (T2b, T3b, T4b), lymph node-negative (N0), and no distant metastasis (M0)

Interventions
  • Experimental: PEG IFN alpha 2b (3 µg/kg weekly injections)
  • No intervention: observation

OutcomesPrimary outcomes of the trial

  1. Relapse-free survival


Secondary outcomes of the trial

  1. Toxicity
  2. Overall survival
  3. Distant metastases-free survival
  4. Quality of life

Starting dateOctober 2012

Contact informationAlexander Eggermont, Institut Gustave Roussy, Paris, France

NotesExpected primary completion date: April 2020

NCT01782508

Trial name or titleA Phase II Randomized Study of Imatinib Versus High Dose Interferon as Adjuvant Therapy in KIT-mutated Patients With Resected Melanoma

MethodsThis is a phase II randomised study

Participants
  • Melanoma patients who had lymph node metastasis removed

Interventions
  • Experimental: Participants will take 400 mg imatinib once daily for 1 year
  • Active comparator: Participants will receive interferon 1500 wiu/m² d1-5 for 4 weeks followed by 900 wiu IH 3 days/week for 11 months

OutcomesPrimary outcomes of the trial

  1. Relapse-free survival


Secondary outcomes of the trial

  1. Overall survival

Starting dateAugust 2012

Contact informationJun Guo, Peking University Cancer Hospital

NotesExpected primary completion date: December 2014

 
Comparison 1. Interferon alpha versus any other comparator

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Disease-free survival (DFS)17Hazard Ratio (Fixed, 95% CI)0.83 [0.78, 0.87]

 2 Overall Survival (OS)15Hazard Ratio (Fixed, 95% CI)0.91 [0.85, 0.97]

 
Summary of findings for the main comparison.

Interferon alpha compared with treatment other than interferon (including observation) for the adjuvant treatment of melanoma

Patient or population: high-risk melanoma participants

Settings: adjuvant

Intervention: interferon alpha

Comparison: observation, treatments other than interferon, or both

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
Number of Participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

Observation or treatmentInterferon alpha

First recurrence50/10044/100HR 0.83 (0.78 to 0.87)10,345 (17 studies)High-qualityFurther research may provide information regarding patient selection and interferon schedule

Death40/10037/100HR 0.91(0.85 to 0.97)9927 (15 studies)High-qualityFurther research may provide information regarding patient selection and interferon schedule

*Assumed risk: For disease-free survival outcome: 5-year disease recurrence rate = 50%; for overall survival outcome: 5-year death rate = 40% (in patients with TNM stage II-III). 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; 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

 
Table 1. Glossary of terms

Medical termExplanation

Adjuvant treatmentAny medical oncology therapy used after surgery to kill microscopic tumour residues not removed by the surgeon

AJCC TNM stages I-IVStage I and II = primary melanomas with negative regional lymph nodes

Stage III = regional lymph node melanoma metastasis from known or unknown primary tumours

Stage IV = melanoma metastasis at distant site (e.g. lung, liver, brain)

ApoptosisApoptosis is a genetically determined process of programmed cell death that may occur in cells. Apoptosis is a normal physiological process eliminating DNA-damaged, superfluous, or unwanted cells and when halted may result in uncontrolled cell growth and tumour formation

AngiogenesisDevelopment of new blood vessels. It may occur in the healthy body for healing of wounds and restoring blood flow to tissues after injury, or in tumours, where it promotes the spread of cancer cells through new blood vessels

LymphadenectomyThis surgical operation aims to remove the lymph nodes of 1 or more of the 3 main fields (neck, axilla, groin) where metastatic melanoma cells are present

MetastasisThe spread of a malignant tumour from its original site to any part of the body (such as lymph nodes, lungs, liver, brain, bones, and so on)

Neoadjuvant treatmentAny medical oncology therapy used before surgery to reduce the tumour bulk

Pegylated interferonThe addition of polyethylene glycol to the interferon molecule to improve the effectiveness of the drug

Randomised controlled trialThis is a particular design of study, in which participants are randomly assigned to different treatments. A randomised controlled trial provides the highest evidence for the use, or not, of a diagnostic or therapeutic intervention

Sentinel node biopsySurgical procedure to find the first lymph node that drains the skin area of the primary melanoma. The pathological evaluation of the sentinel node allows definition of the status of the lymphatic field when no clinically evident regional lymph node metastasis is present

TNM classificationThis is the international classification of tumour spread issued by the American Joint Committee on Cancer (AJCC): "T" refers to the size of a tumour, "N" to the presence and extent of lymph node metastasis, and "M" indicates whether or not distant metastatic disease (that is, to lungs, liver, brain, bones) is present

E.g. AJCC stage II (T2-4N0M0): T2-4 refers to the size and extent of the primary tumour; N0 refers to no regional lymph node involvement; M0 refers to no metastases

AJCC stage III (TanyN+M0): Tany refers to any size and extent of the primary tumour, including thin (< 1.00 mm) and thick melanomas (> 4.00 mm); N+ refers to any number of positive regional lymph nodes

Grades of toxicityGrade 0 = no adverse event or within normal limits

Grade 1 = mild adverse event

Grade 2 = moderate adverse event

Grade 3 = severe and undesirable adverse event

Grade 4 = life-threatening or disabling adverse event

 
Table 2. Included studies

StudyDesignSample sizeSettingParticipants (AJCC TNM stage)Intervention (interferon schedule)Outcome (P value)

Creagan 1995Phase III RCT264AdjuvantII-III

(T2-4N0M0/TanyN+M0)
IFNa(2a): 20 MU/sqm x 3/week for 4 months (route: i.m.)DFS: not significant

OS: not significant

Kirkwood 1996Phase III RCT287AdjuvantII-III

(T4N0M0/TanyN+M0)
IFNa(2b): 20 MU/sqm x 5/week (1 month, route: i.v.) + 10 MU/sqm x 3/week (48 weeks, route: s.c.)DFS: 0.0013

OS: 0.015

Rusciani 1997Phase III RCT154AdjuvantI-II

(TanyN0M0)
IFNa(2a): 3 MU x 3/week (3 weeks, route: i.m.) x 6 months, followed by a month with no treatment x 3 yearsDFS: not reported

OS: not reported

Pehamberger 1998Phase III RCT311AdjuvantII

(T2-4N0M0)
IFNa(2a): 3 MU x 7/week (3 weeks, route: s.c.) + 3 MU sc x 3/week (12 months, route: s.c.)DFS: 0.02

OS: not evaluated

Grob 1998Phase III RCT499AdjuvantII

(T2-4N0M0)
IFNa(2a): 3 MU x 3/week (18 months, route: s.c.)DFS: 0.033

OS: 0.046

Kirkwood 2000Phase III RCT642AdjuvantII-III

(T4N0M0/TanyN+M0)
IFNa(2b) (high): 20 MU/sqm x 5/week (1 month, route: i.v.) + 10 MU/sqm x 2/week (48 weeks, route: s.c.). IFNa(2b) (low): 3 MU x 2/week (2 years, route: s.c.)DFS (high): 0.03

DFS (low): not significant

OS: not significant

Cameron 2001Phase III RCT96AdjuvantII-III

(T3-4N0M0/TanyN+M0)
IFNa(2b): 3 MU x 3/week (6 months, route: s.c.)DFS: not significant

OS: not significant

Kirkwood 2001Phase III RCT880AdjuvantII-III

(T4N0M0/TanyN+M0)
IFNa(2b): 20 MU/sqm x 5/week (1 month, route: i.v.) + 10 MU/sqm x 2/week (48 weeks, route: s.c.)DFS: 0.0027

OS: 0.0147

Cascinelli 2001Phase III RCT444III

(TanyN+M0)
IFNa(2a): 3 MU x 3/week (36 months, route: s.c.)DFS: not significant

OS: not significant

Kirkwood 2001Phase II RCT107AdjuvantII-III-IV

(stage IV: resectable metastatic disease)
IFNa(2b) (d1): IFNa (from day 1) 20 MU/sqm x 5/week (1 month, route: i.v.) + 10 MU/sqm x 3/week (48 weeks, route: s.c.). IFNa(2b) (d28): IFNa as above (from day 28)DFS (d1): 0.016

DFS (d28): 0.03

OS: not significant

Hancock 2004Phase III RCT674AdjuvantII-III

(T4N0M0/TanyN+M0)
IFNa(2a): 3 MU x 3/week (2 years, route: s.c.)DFS: not significant

OS: not significant

Kleeberg 2004Phase III RCT830AdjuvantII-III

(T3-4N0M0/TanyN+M0)
IFNa(2b): 1 MU every other day (12 months, route: s.c.)DFS: not significant

OS: not significant

Eggermont 2005Phase III RCT1388AdjuvantII-III

(T4N0M0/TanyN+M0)
IFNa(2b) (1 year): 10 MU x 5/week (4 weeks, route: s.c.) + 10 MU x 3/week (12 months, route: s.c.)

IFNa(2b) (2 years): 10 MU x 5/week (4 weeks, route: s.c.) + 5 MU x 3/week (24 months, route: s.c.)
DFS: not significant

OS: not significant

Garbe 2008Phase III RCT444AdjuvantIII

(TanyN+M0)
IFNa(2a): 3 MU x 3/week (2 years, route: s.c.)DFS: 0.018

OS: 0.005

Eggermont 2008Phase III RCT1256AdjuvantIII

(TanyN+M0)
Pegylated IFNa(2b): 6 ug/Kg/week (8 weeks, route: s.c.) + 3 ug/Kg/w (5 years, route: s.c.)DFS: 0.02

OS: not significant

McMasters 2008Phase III RCT774AdjuvantII-III

(T2-4N0M0/TanyM0)*
IFNa(2b): 20 MU/sqm x 5/week (1 month, route: i.v.) + 10 MU/sqm x 3/week (48 weeks, route: s.c.)DFS: not significant

OS: not significant

Hansson 2011Phase III RCT855AdjuvantII–III

(T4N0M0/TanyN+M0)
IFNa (2b) (1 year): 10 MU x 5/week (4 weeks, route: s.c.) + 10 MU x 3/week (12 months, route: s.c.)

IFNa (2b) (2 years): 10 MU x 5/week (4 weeks, route: s.c.) + 10 MU x 3/week (24 months, route: s.c.)
DFS: 0.034

OS: not significant

Agarwala 2011Phase III RCT1150AdjuvantII-III

(T3-4N0M0/TanyN1a-N2aM0)
IFNa(2b): 20MU/sqm x 5/week (1 month, route: i.v.)DFS: not significant

OS: not significant

 *Negative sentinel lymph node after standard pathological evaluation (H&E and immunohistochemistry) were tested with polymerase chain reaction (PCR) analysis
 
Table 3. Additional analyses: disease-free survival (DFS)

Trial featureRCTHR (95% CI) #Z-test P valueI² statisticQ-test P valueType of analysisReferences

Interferon dose: high8*0.80 (0.74 to 0.87)< 0.000114%0.32Subgroup meta-analysis1 to 8

Interferon dose: low8*0.85 (0.77 to 0.94)0.00125%0.23Subgroup meta-analysis8 to 15

Interferon dose: intermediate20.84 (0.75 to 0.95)0.0050%0.43Subgroup meta-analysis16, 17

Interferon dose17Q-value = 0.99-0%0.61Heterogeneity1 to 17

 

TNM stage: II20.70 (0.55 to 0.88)0.0020%0.46Subgroup meta-analysis12, 15

TNM stage: III50.85 (0.77 to 0.94)0.0010%0.61Subgroup meta-analysis1, 3, 6, 10, 11

TNM stage: II-III100.83 (0.77 to 0.89)< 0.000132%0.15Subgroup meta-analysis2, 4, 5, 7 to 9, 13, 14, 16, 17

TNM stage17Q-value = 2.23-10%0.33Heterogeneity1 to 17

 

Publication year17Slope = 0.01 (-0.0004 to 0.02)0.06--Metaregression1 to 17

Treatment duration17Slope = 0.0001 (-0.003 to 0.003)0.09--Metaregression1 to 17

 Statistics : Z-test: tests the significance of meta-analysis summary effect;I² statistic: measures between-study heterogeneity in meta-analysis; Q-test: tests the significance of heterogeneity; Q-value: Q-test statistic; slope: coefficient of metaregression model
Abbreviations : interferon: interferon alpha; RCT: randomised controlled trial; HR: hazard ratio (from meta-analysis); CI: hazard ratio confidence interval
References : 1: Agarwala 2011; 2: Creagan 1995; 3: Eggermont 2008; 4: Kirkwood 1996; 5: Kirkwood 2001; 6: McMasters 2008; 7: Kirkwood 2001a; 8: Kirkwood 2000; 9: Cameron 2001; 10: Cascinelli 2001; 11: Garbe 2008; 12: Grob 2010; 13: Hancock 2004; 14: Kleeberg 2004; 15: Pehamberger 1998; 16: Eggermont 2005; 17: Hansson 2011
Notes : *: The starred values indicate that one study (Kirkwood 2000) is represented twice (in both high- and low-dose interferon groups) because this trial had three arms: observation, high-dose interferon, and low-dose interferon. #: The third column displays hazard ratios if not otherwise specified
 
Table 4. Additional analyses: overall survival (OS)

FeatureRCTHR (95% CI) #Z-test P valueI² statisticQ-test P valueType of analysisReferences

Interferon dose: high7*0.93 (0.84 to 1.03)0.1611%0.34Subgroup meta-analysis1 to 7

Interferon dose: low7*0.88 (0.79 to 0.98)0.0223%0.25Subgroup meta-analysis7 to 13

Interferon dose: intermediate20.91 (0.80 to 1.04)0.160%1.00Subgroup meta-analysis14, 15

Interferon dose15Q-value = 0.53-0%0.76Heterogeneity1 to 15

 

TNM stage: II10.70 (0.50 to 0.98)---N/A11

TNM stage: III 50.95 (0.85 to 1.05)0.3243%0.13Subgroup meta-analysis1, 3, 6, 9, 10

TNM stage: II-III 90.90 (0.83 to 0.98)0.010%0.77Subgroup meta-analysis2, 4, 5, 7, 8, 12 to 15

TNM stage15Q-value = 3.11-36%0.21Heterogeneity1 to 15

 

Publication year15Slope = 0.01 (-0.004 to 0.023)0.16--Metaregression1 to 15

Treatment duration15Slope = 0.001 (-0.002 to 0.005)0.52--Metaregression1 to 15

 Statistics : Z-test: tests the significance of meta-analysis summary effect;I² statistic: measures between-study heterogeneity in meta-analysis; Q-test: tests the significance of heterogeneity; Q-value: Q-test statistic; slope: coefficient of metaregression model
Abbreviations : interferon: interferon alpha; RCT: randomised controlled trial; HR: hazard ratio (from meta-analysis); CI: hazard ratio confidence interval; N/A: not applicable
References : 1: Agarwala 2011; 2: Creagan 1995; 3: Eggermont 2008; 4: Kirkwood 1996; 5: Kirkwood 2001; 6: McMasters 2008; 7: Kirkwood 2000; 8: Cameron 2001; 9: Cascinelli 2001; 10: Garbe 2008; 11: Grob 2010; 12: Hancock 2004; 13: Kleeberg 2004; 14: Eggermont 2005; 15: Hansson 2011
Notes : *: The starred values indicate that one study (Kirkwood 2000) is represented twice (in both high- and low-dose interferon groups) because this trial had three arms: observation, high-dose interferon, and low-dose interferon. #: The third column displays hazard ratios if not otherwise specified
 
Table 5. Grade 3 to 4 adverse events after treatment with adjuvant interferon alpha

StudyArmFeverFatigueMyalgiaArthralgiaAnorexiaDizzinessHeadacheMood








Grade 3 (%)Grade 4 (%)Grade 3 (%)Grade 4 (%)Grade 3 (%)Grade 4 (%)Grade 3 (%)Grade 4 (%)Grade 3 (%)Grade 4 (%)Grade 3 (%)Grade 4 (%)Grade 3 (%)Grade 4 (%)Grade 3 (%)Grade 4 (%)

Grob 1998IFN< 10NRNRNRNRNRNRNRNR0NR1NR< 1< 1

ObservationNRNRNRNRNRNRNRNRNRNRNRNRNRNRNRNR

Kirkwood 2000High-dose IFNNRNR231161NRNRNRNRNRNRNRNR81

Low-dose IFNNRNR3081NRNRNRNRNRNRNRNR20

ObservationNRNR0000NRNRNRNRNRNRNRNR00

Kirkwood 2001IFNNRNR20.60.33.80NRNRNRNRNRNRNRNR8.61.3

GSKNRNR1.200.50NRNRNRNRNRNRNRNR0.50.2

Hancock 2004IFN106.70NRNRNRNRNRNRNRNRNRNR3.10

Observation001.30NRNRNRNRNRNRNRNRNRNR1.60

Eggermont 200513-month IFN611417161614151102

25-month IFN81121212161415191

Observation0020202110212021

Garbe 2008IFN00NRNRNRNRNRNRNRNRNRNRNRNR1.60.8

IFN + DTIC00NRNRNRNRNRNRNRNRNRNRNRNR1.60

ObservationNRNRNRNRNRNRNRNRNRNRNRNRNRNRNRNR

Eggermont 2008Peg-IFN4115141NRNRNRNRNRNR4161

Observation001010NRNRNRNRNRNR0011

Hansson 201113-month IFN1.10.49.8NR5.3NR2.8NR3.5NANRNR3.5NR4.91.1

25-month IFN1011.2NR4.9NR4.6NR3.5NANRNR3.1NR2.40

Observation0.401.8NR1.1NR1.4NR0NANRNR0.4NR0.40

 NR: not reported
Not available for Creagan 1995 (grades 3 and 4 not reported), Kirkwood 1996 (adverse events grouped as "constitutional symptoms"), Pehamberger 1998 (grades 3 and 4 not reported), Cameron 2001, Cascinelli 2001, Kirkwood 2001 (E2696), Kleeberg 2004, McMaster 2008, Agarwala 2011