Types of studies

Apart from the intervention/control, participants in the intervention and control groups must have been intended to receive identical care. Only studies on the principle of Mendelian randomization will be considered for this review. This is named after Gregor Mendel (1822–1884), a Moravian and an Augustinian monk. In the haematological context, for a patient’s sibling to be a suitable donor, he/she must have inherited the same tissue type as the patient from their mother and father. Since the chances of there being a match depend on the random assortment of genes at fertilization, only one in four siblings will be expected to have the same tissue type as the patient. Thus, whether or not a patient has a matched sibling donor available is essentially for a random process and the presence or absence of a donor can be used as a surrogate for randomization. Mendelian randomization is a method of overcoming this lack of randomized evidence in standard randomized controlled trials Wheatley 2004.

Any published or unpublished genetically randomized trials will be eligible for inclusion in the review. We will include both full text publications and abstract publications. Due to the need of Mendelian randomization in an allogeneic stem cell transplantation trial only sibling donor versus no donor trials will be included. The design of quasi‐randomized trials is considered to be of low quality leading to unreliable study results. Cross‐over trials are generally considered to be inappropriate if the primary outcome is irreversible (for example, mortality) (Higgins 2008a). If randomized controlled clinical trials are missing, applicable publicated results will be discussed.

Types of participants

We will include adult patients (age >= 18 years) with a confirmed diagnosis of advanced primary cutaneous lymphoma, without gender or ethnicity restriction. We will consider advanced stages of cutaneous T cell lymphoma (tumor stage, WHO/EORTC stage > IIB), including newly diagnosed patients, and those with relapsed or resistant disease.  

Types of interventions

The main intervention will be allogeneic stem cell transplantation compared to conventional chemotherapy or immunotherapy. We will consider any chemotherapeutic and immuno‐chemotherapeutic regimen for comparison.

The intervention is defined as follows.

• Allogeneic stem cell transplantation plus conditioning therapy regardless of agents, with or without radiotherapy, reduced‐intensity conditioning (RIC) or non‐RIC as treatment for advanced cutaneous T cell lymphoma.

The control intervention is defined as follows.

• Conventional therapy for advanced cutaneous T cell lymphoma therapy such as radiation, photopheresis, uv‐irradiation, monoclonal antibodies (e.g. Alemtuzumab), interferon‐alpha, cytostatic monotherapies (e.g. Gemcitabine or liposomal Doxorubicine) or polychemotherapies (e.g. CHOP).

Types of outcome measures

Electronic searches

The search will cover:

• the Cochrane Central Register of Controlled Trials (CENTRAL);

• MEDLINE (see Appendix 1 for search strategy);

We will also search databases of ongoing trials:

• www.controlled‐trials.com/mrct/;

Searching other resources

We will handsearch:

• citations from identified trials and relevant review articles;

• the following conference proceedings:

• • ASH (American Society of Hematology) 2009 to present;

  • ASCO (American Society of Clinical Oncology) 2009 to present.

(Randomised controlled trials from ASH, ASCO, EBMT and International Conference on Cutaneous T Cell Lymphoma up to 2009 were handsearched by the Cochrane Haematological Malignancies Group (CHMG) and are included in CENTRAL).

We will also contact groups or individuals who may have conducted randomized trials in allo SCT as therapy for primary cutaneous T cell lymphoma.

Selection of studies

Two review authors will independently screen titles and abstracts of studies identified from the above sources. The first screening will discard studies that are clearly ineligible. If this cannot be done satisfactorily from the title and abstract, we will obtain a full‐text version and discuss eligibility. The aim is to be overly inclusive rather than to risk losing relevant studies. We will assess selected studies with an eligibility form to determine whether they meet the inclusion criteria. We will resolve any disagreement by discussion. If necessary, we will seek further information from the authors where articles contain insufficient data to make a decision about eligibility. The eligibility form contains the following questions.

1. Is the study described as genetically randomized in order to avoid bias?

2. Is the diagnosis of CTCL histologically confirmed?

3. Were the participants in the experimental group treated by allogeneic stem cell transplantation?

4. Were the participants in the control group treated by chemotherapy or immuno‐chemotherapy or immunotherapy?

5. Due to the need of Mendelian randomization in an allogeneic stem cell transplantation trial only sibling donor versus no donor trials will be included.

Data extraction and management

Two authors will independently extract data concerning details of study population, intervention and outcomes using a standardized data extraction form. This form will include at least the following terms:

• general information: author, title, source, publication date, country, language, duplicate publications;

• study characteristics: trial design, aims, setting and dates, source of participants, examination of proportion of the donor group (Mendelian randomization), inclusion/exclusion criteria, treatment allocation, comparability of groups, subgroup analysis, statistical methods, power calculations, compliance with assigned treatment, length of follow‐up, time‐point of randomization (Mendelian);

• participant characteristics: age, gender, ethnicity, number of participants recruited/allocated/evaluated, participants lost to follow up, additional diagnoses, percentage actually receiving transplant;

• interventions: setting, type of (multi‐agent) chemotherapy (intensity of induction and conditioning regimen, number of cycles, with or without radiation), type and dosage of immunotherapy, type and dosage of antibodies, transplantation with or without growth factor support, transplant details, infection prophylaxis, type of maintenance treatment, type of salvage treatment, duration of follow up;

• outcomes: overall survival, progression‐free survival, response rate, treatment‐related mortality, adverse events, quality of life.

Where possible, we will seek missing data from the authors.

Assessment of risk of bias in included studies

Two review authors will independently evaluate all included trials using a list of selected quality criteria. We will evaluate whether the criterion is fulfilled on a three‐point scale: agree, uncertain, disagree. The criteria will be according to the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2008).

1. Prospective donor availability reporting

2. Allocation concealment

3. Blinding of participants, personnel and outcome assessors

4. Incomplete outcome data

5. Selective outcome reporting

6. Other sources of bias

We will resolve disagreement between the rating review authors by consensus. The review authors will not be blinded to the names of authors, institutions, journals or the outcomes of the trials.

Measures of treatment effect

For binary outcomes, we will calculate risk ratios (RR) with 95% confidence intervals (CI) for each trial. We will calculate time‐to‐event data as Hazard Ratios (HR) according to Parmar (Parmar 1998; Tierney 2007).

We will assess heterogeneity of treatment effects between trials using a Chi² test with a significance level at P < 0.1. We will use statistic to quantify possible heterogeneity (I² > 30% moderate heterogeneity, I² > 75% considerable heterogeneity) (Deeks 2008).

Dealing with missing data

We will impute missing continuous data where necessary (e.g. calculating standard deviations (SDs) from standard errors and P values) and clearly document this in the review. Where it is not possible to calculate missing SDs, we will carry out available case analysis based on participants who completed a trial (Higgins 2008).

Assessment of heterogeneity

We will inspect heterogeneity (the degree of difference between the results of different trials) graphically in the results of the trials and assess this by calculating a test of heterogeneity (Chi² and I²). We will explore potential causes of heterogeneity by sensitivity and subgroup analysis using meta‐regression (Deeks 2008).

Assessment of reporting biases

In meta‐analyses with at least 10 trials we will examine a funnel plot estimating the treatment effect against the precision of trials (plots of the log of the relative risk for efficacy against the standard error) in order to estimate potential asymmetry that may indicate selection bias (the selective publication of trials with positive findings) or methodological flaws in the small studies. We will also estimate publication bias by the formal linear regression test (Egger 1997, Sterne 2008). We will consider a P value of less than 0.1 significant for this test.

Data synthesis

We will perform analyses according to the recommendations of The Cochrane Collaboration. We will use the Cochrane software package Review Manager 5 RevMan2008 for analysis.

We will pool HRs and their confidence intervals for time‐to‐event outcomes using an inverse variance method, and RRs and their confidence intervals for dichotomous data using the Mantel‐Haenszel method. We will use a fixed‐effect model. We will repeat primary analyses using a random‐effects model (DerSimonian and Laird method) for sensitivity analyses.

Subgroup analysis and investigation of heterogeneity

We will consider the following subgroups:

• age (adults < 65 years, adults >= 65 years);

• stage;

• type and intensity of preparative regimen;

• antibody usage;

• immunotherapy (interferon) usage;

• patients newly diagnosed or with relapse;

• duration of follow up;

• RIC versus non‐RIC (conditioning therapy)

We will assess heterogeneity of treatment effects between trials by using a Chi² test with a significance level at P < 0.1. We will use the I² statistic to quantify possible heterogeneity.

Sensitivity analysis

We will base sensitivity analyses on:

• quality components, including full text publications/abstracts, preliminary results versus mature results;

• fixed‐effects versus random‐effects model.