Neoadjuvant treatment for malignant and metastatic cutaneous melanoma
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To assess the effects of neoadjuvant treatments in adults with American Joint Committee on Cancer (AJCC) stage III or stage IV melanoma.
Background
Description of the condition
Cutaneous melanoma is amongst the most aggressive of all skin cancers (Garbe 2016). It is a type of skin cancer originating in the melanin‐producing melanocytes, which are found between the outer layer of the skin (the epidermis) and the layer beneath (the dermis); melanomas can also arise on the eye, meninges, and mucosal surfaces (Garbe 2016). Cutaneous melanoma typically presents in distinct subtypes, including superficial spreading melanoma (which remains in a horizontal growth phase until vertical invasion and spread occurs), nodular melanoma (which appears as a nodule from the outset), acral lentiginous melanoma (which is present on acral surfaces such as the sole, and is more common in Asian populations), and lentigo maligna melanoma (invasive melanoma developing in a premalignant condition called lentigo maligna).
Melanocytes can become cancerous as a result of unrepaired DNA damage or other genetic alterations, or both (Curtin 2005; Eggermont 2014). There are several genetic and environmental factors that increase the risk of melanoma, including exposure to sunlight and ultraviolet (UV) radiation; having a high number of moles (naevi); being very fair skinned (especially with fair or red hair); family history; age; and having a history of previous melanoma (Whiteman 2011).
Melanoma occurs primarily in European, North American, and Oceanic populations, which account for almost 82% of the global incidence of melanoma (Ferlay 2015). These populations also represent approximately 64% of mortality associated with the disease (Ferlay 2015). In 2012, there was a global incidence of 232,130 cases (1.7% of total cancer cases) and 55,488 deaths (0.7% of total cancer mortality) (Ferlay 2015). The Lancet Global Burden of Disease study showed that the incidence of malignant melanoma has increased between the years 2006 and 2016 and the age‐standardised death rate and years of life lost due to malignant melanoma decreased at a lower rate than the average for all neoplasms (Global Burden of Disease 2016). Models predict this global trend of increasing incidence will continue (Garbe 2009), although there are signs that incidence may have peaked in Australia and New Zealand, the countries with the highest incidence in the world (Whiteman 2016). Incidence and mortality are higher in men than women (Ferlay 2015).
Dermoscopy in conjunction with histopathology assessment is commonly used to diagnose and stage melanomas; sentinel lymph node biopsy can also be used for staging (Garbe 2016). Melanoma is staged according to the American Joint Committee on Cancer (AJCC) Melanoma Staging criteria (Balch 2009). In stage 0 melanoma (in situ melanoma), the abnormal melanocytes have not started to spread into deeper layers. In stage I and II melanoma, an invasive cancer has formed, but there is no spread to lymph nodes or distant sites. With stage III melanoma, the melanoma has spread to the lymph nodes or lymphatic channels and it may or may not be ulcerated. In stage IV melanoma, the cancer has spread elsewhere in the body, with the brain, lung, liver, the distant lymph nodes and other areas of the skin being the most common places of metastasis (Dummer 2015). Prognostic variables in melanoma include mitotic rate, number and site of distant metastases, and serum lactate dehydrogenase (LDH) levels (based on observational research) (Balch 2009). Male gender, age, and site of primary melanoma have also been identified as relevant through observational research (Garbe 2016). Tumour‐infiltrating lymphocytes (white blood cells that migrate into a tumour and help kill tumour cells as part of the host immune response to cancer) have also been identified as a potential prognostic factor in melanoma, at both early and advanced stages of the disease (Thomas 2013). Ten‐year survival ranges from 93% for stage IA to 39% for stage IIC; and is markedly worse for stage III, where five‐year survival rates range from 78% to 40%. The five‐year survival in stage IV disease ranges from 10% to 25%, depending on LDH levels (Balch 2009).
Melanoma tumours are known to have an exceptionally high mutational load due to the combination of driver genetic mutations and continuous exposure to the carcinogen, UV radiation (Curtin 2005). Recent work by the Cancer Genome Atlas suggested classification of melanomas according to four genomic categories: BRAF, RAS, NF1 (neurofibromin 1), and Triple‐Wild Type (Akbani 2015). Currently available treatments target the BRAF mutation, and research is ongoing to identify therapeutic agents which target the other mutations (Posch 2013). The high mutational load and immunogenicity of melanoma tumours has contributed to melanoma being the first disease area where checkpoint inhibitors were investigated and received marketing authorisation in the USA and Europe (Postow 2015).
Description of the intervention
Neoadjuvant treatment is a form of induction therapy, given as a first step to shrink a cancerous tumour prior to the main treatment, which is generally surgery (NCI 2017). The aim of neoadjuvant treatment in melanoma can be to improve survival outcomes, shrink tumours to make them more amenable to surgical intervention, reduce surgical morbidity, or to improve surgical outcomes (Tahrini 2011). Neoadjuvant treatment is generally administered for a planned, fixed period of time prior to a surgical procedure, and then may or may not continue to be administered in the postoperative period. There are various approaches which can be employed in the neoadjuvant treatment of tumours, including cytotoxic chemotherapy, radiotherapy, topical agents, immunotherapies, and targeted treatments. These agents work by different pharmacological and physiological mechanisms to reduce tumour volume, making surgical intervention more feasible.
A retrospective analysis of data from the Surveillance, Epidemiology and End Results (SEER) database in the USA indicated that metastectomy (surgical removal of metastases) was a significant predictor of survival in patients with stage IV melanoma (Wasif 2011). A retrospective observational study comparing survival outcomes in stage IV melanoma in patients undergoing metastectomy and systemic anti‐cancer treatment to those receiving systemic treatment alone, found a clear survival advantage for those patients undergoing surgical intervention in addition to systemic treatment (Howard 2012). It follows that treatments which may facilitate successful metastectomy could be associated with improved survival. However, it should be noted that this study was undertaken before the advent of effective systemic therapies. Neoadjuvant treatment was generally directed at patients with stage III disease, but with the advent of new treatments capable of generating significant tumour shrinkage at multiple sites, there is greater interest in using it to treat loco‐regional disease, to render large tumours or metastases resectable (Tahrini 2011).
Neoadjuvant treatment strategies are not currently included in the European Society of Medical Oncology (ESMO) treatment guidelines for cutaneous melanoma (Dummer 2015). The National Comprehensive Cancer Network (NCCN) guidelines similarly do not recommend any neoadjuvant treatment strategies, but do refer to four currently ongoing trials investigating checkpoint inhibitors, targeted BRAF/MEK combination treatment and intra‐lesional vaccination (Fields 2016). Neoadjuvant treatment is included as a treatment option for distant metastases in the European Consensus Guidelines 2016 issued by the European Organisation for Research and Treatment of Cancer (EORTC) and European Association of Dermato‐Oncology (EADO) (Garbe 2016), referencing research using neoadjuvant treatment with high‐dose interferon (Moschos 2006).
How the intervention might work
A large variety of therapeutic agents have been investigated for the neoadjuvant treatment of melanoma. Evidence suggests that neoadjuvant treatments for melanoma may exert their effects through immuno‐modulatory mechanisms rather than through direct anti‐tumour effects (Moschos 2006; Johnson 2015a). Treatments such as immunotherapies and targeted treatments have demonstrated efficacy in advanced disease, but their utility is constrained by heterogeneity in patient response and the development of tumour resistance over time (Johnson 2015; Zhao 2017). It is thought that these agents may be more effective in earlier stages of disease, prior to changes in the tumour microenvironment which facilitate tumour escape mechanisms, and this forms the underlying therapeutic hypothesis for neoadjuvant therapy (Braeuer 2014; Davar 2013).
Chemotherapy
Cytotoxic chemotherapy was once the backbone of treatment for stage IIIc and IV disease. Numerous regimens have been investigated over the decades, exhibiting variable levels of disease response, but there is no proven impact on overall survival (OS) (Pasquali 2018). Dacarbazine is an alkylating agent (Lexicomp, 20th Ed) which works by disrupting the DNA replication mechanisms of the tumour, and has been investigated as monotherapy for stage IIIc and IV disease, as well as in neoadjuvant and adjuvant treatment strategies for melanoma (Buzaid 1998; Kim 2009). Although now largely displaced by the newer agents, it still plays a role in palliative chemotherapy, and as first‐line treatment where newer treatments are not available or not reimbursed. Response rates for dacarbazine vary from 5% to 15% (Bhatia 2009), and there is no evidence of OS benefit associated with its use (Sasse 2007; Pasquali 2018). Temozolomide is an oral analogue of dacarbazine, and has demonstrated non‐inferiority to dacarbazine in phase III randomised controlled trials (RCTs) (Middleton 2000; Patel 2011). Combination chemotherapy regimens targeting multiple mechanisms of cell growth and replication have also been investigated, using agents including the vinca alkaloids, such as vindesine and vinblastine (inhibitors of microtubular assembly), taxanes such as paclitaxel (inhibitors of microtubule disassembly), platinum analogues such as cisplatin or carboplatin (alkylating agents), and nitrosoureas such as lomustine, carmustine and fotemustine (alkylating agents) (Bhatia 2009). Frequently used combinations included the Dartmouth regimen (cisplatin, dacarbazine, bendamustine and tamoxifen), CVD (cisplatin, vinblastine and dacarbazine), and PC (paclitaxel and carboplatin). Compared to monotherapy, combination regimens are associated with an increase in toxicity, a slightly higher response rate, and no significant improvement in OS (Pasquali 2018).
Immunotherapy
Earliest investigations of immunotherapy in melanoma involved interleukin‐2 (IL‐2) and interferon alpha (IFN‐alpha) in various treatment regimens, for stage III to IV disease and also as adjuvant treatment (Kirkwood 2012). IFN‐alpha is indicated for stage II and III adjuvant treatment, and is associated with increased disease‐free survival (DFS), and potentially an increase in OS (Mocellin 2013). Its precise mechanism of action in melanoma is unknown, although it is postulated to be linked to its immuno‐stimulatory effects on antigen‐presenting cells, producing an increase in tumour‐infiltrating lymphocytes causing an innate immune response to the tumour, rather than a direct anti‐tumour effect (Moschos 2006; Heise 2016). IFN‐alpha has also been investigated in conjunction with chemotherapy (referred to as biochemotherapy), demonstrating improvements in progression free survival (PFS) but not OS, and higher toxicity rates than chemotherapy alone (Pasquali 2018)). Post hoc subgroup analyses have suggested that the greatest effect is seen in patients with ulcerated early‐stage melanoma, and this is currently under investigation in the EORTC 18081 trial (Wheatley 2007; Eggermont 2012). IL‐2 has demonstrated a positive and durable effect on disease response in approximately 10% of patients with advanced disease, but its use is limited by the severe toxicity and lack of biomarker to predict efficacy (Amaria 2015). The administration of IL‐2 has multiple effects at the tumour site and the precise mechanism of action is unclear: it promotes differentiation and proliferation of T lymphocytes, stimulates the production of cytokines, and may increase vascular permeability. It has been displaced as a first‐line treatment option by newer agents, but continues to be used within specialist centres and clinical trials (Amaria 2015).
More recently, the cytotoxic T‐lymphocyte associated protein 4 (CTLA‐4) inhibitor ipilimumab showed OS benefit in stage IIIc and IV melanoma, both as monotherapy and in combination with dacarbazine, compared to a peptide vaccine and dacarbazine alone (Hodi 2010; Robert 2011), and as an adjuvant treatment for stage III disease (Eggermont 2016). It is currently undergoing investigation in the neoadjuvant setting. It is associated with a modest increase in OS of approximately three months and has demonstrated a durable response in a small proportion of patients (Schadendorf 2015). It acts by attenuating inhibitory signals to T cells from dendritic cells, allowing a T‐cell driven response to the tumour to develop; a number of additional local actions have been postulated including reducing T‐regulatory cell function (T‐regulatory cells suppress the immune responses of other cells) and altering T‐cell ratios which drives tumour cell death (Postow 2015).
In 2015 two agents, nivolumab and pembrolizumab, both targeting the programmed death‐1 (PD‐1) regulatory pathway, were licensed in stage IIIc and IV melanoma (Postow 2015). Disruption of the binding of the programmed death‐ligand 1 (PD‐L1) with its PD‐1 receptor potentiates the T‐cell immune response against the tumour, and may influence the activity of other immunologic cells such as B cells and natural killer cells (Postow 2015). Monotherapy with these agents has demonstrated statistically significant improvement in progression free survival (PFS) and OS compared to dacarbazine and ipilimumab (Ribas 2015; Weber 2015; Schachter 2016). The synergistic effects of joint CTLA‐4/PD‐1 inhibition with nivolumab and ipilimumab were confirmed in the pivotal phase II and III RCTs CheckMate 069 (Hodi 2016) and CheckMate 067 (Larkin 2015; Wolchok 2017) demonstrating large gains in PFS and OS compared to ipilimumab alone.
Collectively, ipilimumab, pembrolizumab and nivolumab are known as checkpoint inhibitors, and have become the standard of care treatments for stage IIIc and IV melanoma; all are associated with increases in OS, with a variable and unpredictable disease response, with some subgroups achieving a durable response characterised by a unique immune‐related side effect profile (Postow 2015). Because of the immuno‐stimulatory effects of these treatments, there is widespread research interest in their potential to prevent disease recurrence in earlier stage disease, and various combinations and strategies are under investigation in the adjuvant and neoadjuvant setting, with preliminary evidence of efficacy emerging (Saw 2016).
Early research with granulocyte‐macrophage colony stimulating factor (GM‐CSF) showed some impact on disease response in advanced melanoma (Si 1996; Hoeller 2001; Ridolfi 2002), thought to be mediated by stimulation of dendritic cells to trigger a host immune response. This led to the development of talimogene laherparepvec, an oncolytic viral immunotherapy derived from Herpes Simplex Virus‐1 (HSV‐1), which is designed to produce GM‐CSF intra‐lesionally (Andtbacka 2015). This agent has demonstrated efficacy in the treatment of regionally or distantly metastatic melanoma (stage IIIb, IIIc and IV) in the absence of visceral metastases and normal LDH levels; it was associated with an increase in OS, ORR and time to treatment failure compared to GM‐CSF monotherapy in a phase III RCT (Kaufman 2014). Concomitant administration of sargramostim, a GM‐CSF‐secreting vaccine adjuvant, with ipilimumab also demonstrated improved treatment outcomes in a phase III RCT (Hodi 2014). Talimogene laherparepvec is currently undergoing investigation in neoadjuvant treatment of melanoma, and multiple trials involving sargramostim are ongoing (Andtbacka 2015).
Targeted treatments
The importance of the RAS‐RAF‐MEK‐ERK pathway in melanoma genesis is long established; the BRAF gene, and its protein product BRAF kinase, are key regulators of this pathway (Figure 1; Davies 2002; Eggermont 2014). The BRAF gene is mutated in approximately 50% of cutaneous melanomas and increases cell proliferation and tumour growth (Eggermont 2014); inhibition of these actions can have a detrimental effect on tumour growth. The BRAF inhibitors vemurafenib and dabrafenib were licensed in Europe in 2011 and 2013 respectively, and are associated with statistically significant increases in PFS and OS compared to dacarbazine (Hauschild 2012; McArthur 2014). Treatment resistance quickly emerged as a barrier to long‐term response in almost all patients (Sullivan 2013). Later, the MEK inhibitors trametinib and cobimetinib were licensed for concomitant use with dabrafenib and vemurafenib respectively, exhibiting a synergistic effect in prolonging PFS and OS compared to BRAF inhibitor monotherapy, and overcoming the resistance issues with durable responses in some patients with favourable survival characteristics (Larkin 2014; Robert 2015; Long 2017). Combined treatment with BRAF/MEK inhibitors is currently under investigation in the adjuvant and neoadjuvant treatment setting for melanoma (Saw 2016; Long 2017a), and also in other solid tumours with the BRAF mutation, such as lung cancer (Planchard 2016) and colorectal cancer (Corcoran 2015).
Simplified diagram of the RAS‐RAF‐MEK‐ERK pathway
Bevacizumab is an anti‐vascular epithelial growth factor (VEGF) monoclonal antibody. It is an anti‐angiogenic agent that exerts its effects by reducing the growth of blood vessels required by growing tumours. It has shown some effect on PFS and OS in a number of solid tumours such as ovarian and colorectal cancer (Giantonio 2007; Perren 2011). Studies have shown promising activity in melanoma (Varker 2007; Kim 2012; Kruijff 2012), and a phase III RCT in the adjuvant setting has shown an increase in disease‐free survival but not demonstrated effect on OS (Corrie 2017). Axitinib is an oral anti‐VEGF agent, which exerts its effects in a similar manner to bevacizumab, and is primarily used in renal cell carcinoma. It has produced both complete and partial responses in patients with previously treated metastatic melanoma (Fruehauf 2011; Algazi 2015).
Topical agents
Imiquimod is a toll‐like receptor (TLR) 7 agonist which acts as an immune response modifier, although its precise mechanism of action is far from clear (Lexicomp, 20th Ed). It is currently used for the topical treatment of superficial basal cell carcinoma, in addition to a number of other indications including genital warts and actinic keratosis (EMA 2016). There is a number of documented case series of its use for the treatment of melanoma, in particular for patients with multiple cutaneous in‐transit metastases (Florin 2012).
Radiotherapy
Radiation therapy uses high‐energy radiation to shrink tumours and kill cancer cells by damaging their DNA so that they can no longer replicate. The role of radiotherapy (RT) in the management of melanoma has traditionally been peripheral, focused primarily in the management of brain metastases (stereotactic ablative RT) and symptom control. For the treatment of localised disease, RT is considered after resection of bulky nodal disease; RT reduces the likelihood of recurrence in the radiation field with no effect on DFS and OS (Dummer 2015). Preclinical models have shown a potential synergistic effect of RT with immunotherapy, with some clinical evidence for the abscopal effect and many case studies and case series reported (Barker 2014;Chandra 2015), although the underlying molecular mechanisms of this effect are poorly understood (Reynders 2015). Clinical trials are underway which are investigating the concomitant use of various dosing schedules of RT with immunotherapy for systemic treatment of advanced disease (Kang 2016).
Why it is important to do this review
Neoadjuvant treatment strategies have proven to be successful in a number of solid tumours including breast, oesophageal and ovarian (van Hagen 2012). While not universally implemented as a treatment strategy in the current treatment paradigms for melanoma, there has historically been interest in this area. Neoadjuvant treatment is a suggested option in the 2016 European consensus guidelines for the management of distant metastases of melanoma (Garbe 2016). In the absence of drug treatments licensed in the neoadjuvant setting, it is important to uncover and examine the underlying evidence base for neoadjuvant treatment recommendations.
With recent therapeutic advances in the systemic treatment of stage IIIc and IV melanoma, there is active research interest in the possibility of utilising these new agents in earlier stages of the disease. In order to assess the benefit of newer agents, it is important to systematically analyse the evidence on benefits of neoadjuvant treatments used for melanoma. There is no published high‐quality systematic review of the trials investigating neoadjuvant treatment strategies for melanoma. This review will provide physicians, researchers and patients with a systematic evaluation of the current evidence base for neoadjuvant treatment, and will serve to provide comparative evidence for the relative efficacy of neoadjuvant treatment and a new generation of therapies.
Objectives
To assess the effects of neoadjuvant treatments in adults with American Joint Committee on Cancer (AJCC) stage III or stage IV melanoma.
Methods
Criteria for considering studies for this review
Types of studies
We will include only randomised controlled trials (RCTs) investigating neoadjuvant treatment approaches for cutaneous melanoma, in patients with malignant or metastatic melanoma (AJCC stage III or IV). We will include only studies that prospectively identify participants. Cluster‐randomised trials will be eligible for inclusion. Non‐randomised studies and cross‐over studies will be excluded from the review of treatment effects. The review will consider only health economics studies conducted alongside effectiveness studies included in the effectiveness component of the review.
Types of participants
We will include studies whose participants have AJCC stage III and IV cutaneous melanoma and are in receipt of neoadjuvant treatment.
This patient population is historically associated with poor survival outcomes and therefore is the most likely to receive treatment interventions. Patients with AJCC stage I and II disease will be excluded from the review as most will undergo surgical resection without subsequent treatment.
The AJCC staging system for melanoma has been in use for many years and is unlikely to result in the exclusion of relevant studies. We will include studies with participants of any age, gender, or ethnicity. Studies with only a subset of the proscribed patient population (e.g. stage III but not stage IV) will be included if they meet all other inclusion and exclusion criteria, and the study author can provide separate data for the subset of participants. Similiarly, if we identify a study with a mixed population of stage II and III patients, we will contact the author to obtain subgroup analyses of the specified outcomes for the stage III patients only.
We will exclude trials investigating treatments in lentigo maligna, uveal, and mucosal melanomas, due to differences in disease pathogenesis.
For a study to be included in this review, the neoadjuvant treatment strategy must be clearly specified and meet the following criteria:
-
confirmed disease stage in accordance with the AJCC criteria;
-
predefined systemic treatment prior to planned surgical procedure;
-
planned surgical procedure;
-
may or may not include continued treatment beyond surgical procedure.
Types of interventions
We will consider all types of systemic therapies, radiotherapy or topical drug therapy for the neoadjuvant treatment of stage III and IV melanoma, including:
-
chemotherapy;
-
immunotherapy;
-
targeted treatments;
-
topical agents;
-
radiotherapy.
We will include combinations of the named interventions and regard them as a separate treatment strategy, in line with current therapeutic developments in the metastatic treatment paradigm. Any treatment schedule (i.e. sequence, doses, combinations etc.) will be considered, as long as it meets the defined criteria for neoadjuvant treatment.
Control or comparator arms will include any of the above listed treatments, placebo, standard care or no treatment/observation.
Types of outcome measures
Primary outcomes
-
Overall survival, expressed as a hazard ratio (HR).
-
Adverse events (AEs), expressed as the proportion of patients with Grade three or four AEs on the Common Terminology Criteria for Adverse Events (CTCAE) scale (CTCAE 2009).
Secondary outcomes
-
Overall Response Rate (ORR), expressed as the percentage of patients showing complete response (CR), partial response (PR), stable disease (SD) or progressive disease (PD).
-
Progression‐free survival, expressed as a HR.
-
Disease‐free survival, expressed as a HR.
-
Quality of life, as defined by the validated quality‐of‐life measures or instruments used in each trial.
-
Economic evaluation will be described, expressed as the cost/Quality Adjusted Life Year (QALY) and cost/Life Year Gained (LYG).
-
Pathological complete response (pCR) rate, expressed as the rate of participants showing an absence of residual invasive and in situ cancer on hematoxylin and eosin evaluation of the complete resected specimen and all sampled regional lymph nodes following completion of neoadjuvant systemic therapy.
-
Surgical outcomes (qualitative description as there is not an established measure of surgical outcomes available).
-
Time to recurrent disease, expressed as a HR.
If dichotomous or continuous outcomes are reported at multiple time points, we will report outcomes at six months, one year, two years, and five years where possible.
We will include studies regardless of whether they quantify the prespecified outcomes or not.
Currently there is no defined outcome set for clinical trials in cutaneous melanoma (COMET Initiative 2017). Selection of appropriate endpoints for neoadjuvant trials is an ongoing regulatory challenge and there is no evidence in the literature of a standardised endpoint emerging. Rate of pCR was accepted as an appropriate endpoint in the regulatory approval of neoadjuvant pertuzumab for breast cancer (FDA 2014; Prowell 2012), although its relevance in the clinical setting is unclear (Gnant 2016).
'Summary of findings' tables and GRADE assessments
Using the GRADE approach, we will assess the certainty of evidence for the following outcomes: overall survival, adverse events, overall response rate, progression‐free survival and quality of life. This process involves assessing the certainty of the evidence according to the risk of bias, inconsistency, indirectness, imprecision, and publication bias (GRADE Handbook 2013). Two authors (CG, HOD) will independently assess the certainty of the evidence, and a third (LMcC) will resolve any disputes. Evidence will be graded as high, moderate, low or very low. We will use GRADEpro software to manage the GRADE assessment process and produce 'Summary of findings' tables (GRADEpro GDT). We will present a 'Summary of findings' table for each comparison. If possible, we will report outcomes according to disease stage.
Search methods for identification of studies
We aim to identify all relevant RCTs regardless of language or publication status (published, unpublished, in press or in progress).
Electronic searches
We will search the following databases for relevant trials:
-
the Cochrane Skin Group specialised Register;
-
the Cochrane Central Register of Controlled Trials (CENTRAL) in the Cochrane Library;
-
MEDLINE via Ovid (from 1946 onwards);
-
Embase via Ovid (from 1974 onwards); and
-
LILACS (Latin American and Caribbean Health Science Information Database, from 1982).
We have devised a draft search strategy for RCTs for MEDLINE (Ovid) which is displayed in Appendix 1. This will be used as the basis for search strategies for the other databases listed.
Searching other resources
Trials Registries
We will search the following trial registries using the search terms ‘melanoma’ and restricting to randomised trials only:
-
the ISRCTN register (www.isrctn.com);
-
ClinicalTrials.gov (www.clinicaltrials.gov);
-
the Australian New Zealand Clinical Trials Registry (www.anzctr.org.au);
-
the World Health Organization International Clinical Trials Registry Platform (ICTRP) (apps.who.int/trialsearch/); and
-
the EU Clinical Trials Register (www.clinicaltrialsregister.eu).
Other resources
Searching reference lists
We will check the bibliographies of included studies for further references to relevant trials.
Searching within other reviews
We will search Embase and MEDLINE to retrieve published systematic reviews related to this review title, so that we can scan their reference lists to identify additional relevant trials.
Searching by contacting relevant individuals or organisations
We will contact experts/organisations in the field to obtain additional information on relevant trials.
Handsearching of conference proceedings
Conference proceedings from ASCO, ESMO and SMR are indexed in Embase up to 2015. We will perform a handsearch of the proceedings from the 2016 and 2017 ASCO meetings, and the 2016 SMR and ESMO meetings.
Unpublished literature
We will contact original authors/investigators for clarification and further data if trial reports are unclear.
Secondary endpoints
We will not perform a separate search for information relating to secondary endpoints including AEs and quality of life data. We will consider data on these outcomes that are contained in included studies only.
Data collection and analysis
Selection of studies
We will use Covidence (Covidence 2017) to assess the titles and abstracts identified through the search. Three authors (CG, HOD and SB), working independently, will assess the relevance of all the identified titles and abstracts identified in the search. The full text of any potentially relevant study will be obtained and reviewed for eligibility. An additional author (LMcC) will arbitrate through discussion in instances where eligibility is disputed. Studies excluded at the full text stage will be listed in the 'Characteristics of excluded studies' table, with reasons for exclusion, in the full review. We will report the search steps and results in accordance with the PRISMA statement (Moher 2009). If additional information is required to ascertain eligibility status, we will contact the authors. We will tabulate details of our efforts to contact authors in the review, including details of the information requested, dates, and replies.
Data extraction and management
Data extraction will be conducted independently and in duplicate by two authors (CG and HOD). A third author (SB) will independently review the extracted data for accuracy, and resolve any disagreements. The authors will not be blinded to the journals, trial authors or institutions were the trials were conducted. We will create and pilot a data extraction form, using a sample of the studies. Where relevant information is missing from the study, we will contact the authors. For multiple reports of the same study, we will extract data from each report separately. No restrictions on the timing of the outcomes assessment is planned, as it is anticipated there will be variation in the timing and administration of the neoadjuvant treatment strategies.
The following data will be extracted:
-
descriptive information on the patient population including patient characteristics and disease stage;
-
trial methods including study start date, duration of follow‐up, and funding source;
-
intervention and comparator details, including treatment name, dose, method of administration, duration of treatment and follow‐up;
-
primary and secondary outcomes as specified above.
For time‐to‐event outcomes, we will extract event rate or HR as relevant. If not reported, we will calculate HRs from Kaplan Meier survival plots using dedicated methods (Parmar 1998). For all other outcomes, we will extract the mean, standard deviation or standard error, and the name of the scale used if relevant.
Assessment of risk of bias in included studies
Risk of bias will be assessed independently by two authors (CG and HOD) using the Cochrane ‘Risk of Bias’ tool according to the methods outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). This tool assesses the risk of bias under the following headings: random sequence generation; allocation concealment; blinding of participants and personnel; blinding of outcome assessment; incomplete outcome data and selective outcome reporting. Assessments of bias will be assigned according to the information provided in the study; studies will be categorised as ‘low risk’, high risk’ or ‘unclear risk’ for each domain, according to the specified criteria in theCochrane Handbook. Disagreements will be resolved by consultation with a third party (LMcC). This information will be included in the review in the ‘Risk of Bias’ table, constructed using Review Manager (Review Manager 2014). If insufficient information is available to assess the risk of bias we will contact the authors to obtain or clarify the required information.
Measures of treatment effect
We will use hazard ratios (HRs) and their corresponding 95% confidence intervals (95% CIs) for time‐to‐event outcomes e.g. disease‐free survival (DFS), progression‐free survival (PFS), overall survival (OS) etc. We will extract HRs directly from the original studies when reported. For continuous outcomes we will calculate mean differences and standard deviations where possible; otherwise we will calculate the standardised mean difference. We will calculate risk ratios (RR) and their corresponding 95% CI for dichotomous outcomes including tumour response rate.
Unit of analysis issues
It is anticipated that included studies will be parallel‐group design, and so the unit of analysis is expected to be at individual participant level. If cluster‐randomised trials are included, we will use the published effect estimates taking clustering into account. If multiple‐arm trials are included, we will take account of within‐study correlation, by calculating an average of the relevant pair‐wise comparisons from the study and calculating a variance for the study, accounting for the correlation between the comparisons (Higgins 2011).
Dealing with missing data
When possible, we will conduct analyses using the intention‐to‐treat (ITT) population. If data are missing we will contact the authors to obtain the relevant information. If the ITT analyses are not available, we will perform the analysis using the ‘as treated’ population. The risk of attrition bias will be assessed by examining dropout rates, withdrawals and loss to follow up, and we will describe and critique any methods employed in the publication to address incomplete date, e.g. imputation. Sensitivity analyses will be performed using best‐case and worst‐case scenario analyses for binary outcomes. In the 'Discussion' section of the review, we will consider the impact of missing data on the findings of the review. If possible, missing standard deviations will be imputed from other reported studies, using the maximum reported values in those studies. In all cases the most conservative assumptions will be made.
Assessment of heterogeneity
An assessment of clinical and methodological heterogeneity will be performed as part of assessment for suitability for meta‐analysis, particularly focusing on disease stage, trial design and patient characteristics as important prognostic factors for outcomes (i.e. clinical and methodological differences between trials). Statistical heterogeneity between trials will be assessed using the I2 statistic. The interpretation of the I2 statistic will be based on the ranges provided in the Cochrane Handbook, as follows:
-
0‐40%: Might not be important
-
30‐60%: May represent moderate heterogeneity
-
50‐90%: May represent substantial heterogeneity
-
75‐100%: Considerable heterogeneity
It is recognised that statistical tests for heterogeneity are not sensitive when few studies are involved, and the threshold may not be appropriate depending on the number of studies returned (Higgins 2011).
Assessment of reporting biases
It is not envisaged that this review will include sufficient studies for a funnel plot to appropriately assess reporting bias. We will present a funnel plot if more than 10 studies are found for any one treatment comparison of the primary outcomes (Higgins 2011). In the event that fewer than 10 studies are reported, we will include a narrative description of the risk of reporting bias for the primary outcomes.
Data synthesis
A meta‐analysis of outcomes will only be undertaken if participants, interventions, comparisons and outcomes are considered sufficiently similar across the identified trials to produce a clinically meaningful result. To account for the likely existence of some level of heterogeneity between studies, a random‐effects model will be employed rather than a fixed‐effect model. However, should we only identify a small number (less than four studies) of trials for a given comparison, we will implement a fixed‐effect model, due to the difficulty of estimating between trial heterogeneity (Higgins 2011). If data synthesis occurs, reports with the most similar duration of follow up will be included in the analysis.
Quality of life and economic evaluation evidence will be assessed as a secondary outcome to this review. No synthesis of this evidence will be conducted, but rather a narrative description of the outcomes in terms of costs per quality‐adjusted life year (QALY), costs per life year gained (LYG), and patient‐reported quality of life information.
Subgroup analysis and investigation of heterogeneity
Changes in the tumour environment as the disease progresses may be an effect modifier. A Cochrane review investigating the use of IFN‐alpha in stage II and III melanoma found differing efficacy according to disease stage (Mocellin 2013). Since disease prognosis is different according to disease stage (Balch 2009), we will conduct subgroup analysis examining the effect of intervention in patients with stage III or stage IV disease, if possible.
Sensitivity analysis
If possible, we will perform a sensitivity analysis excluding studies at high risk of bias, by removing studies with at least one domain rated at high risk of bias, irrespective of the type of bias. If PFS is assessed by both blinded and unblinded assessors, the blinded assessment will be used for the primary analysis, and the unblinded assessments for a sensitivity analysis.
Simplified diagram of the RAS‐RAF‐MEK‐ERK pathway
