Review of survival analyses published in cancer journals

Clinical activity of bortezomib in relapsed/refractory MALT lymphomas: results of a phase II study of the International Extranodal Lymphoma Study Group (IELSG)

BACKGROUND

The nuclear factor-kappa B activation in mucosa-associated lymphoid tissue (MALT) lymphoma pathogenesis provided the rationale for the evaluation of bortezomib in this malignancy.

PATIENTS AND METHODS

Thirty-two patients with relapsed/refractory MALT lymphoma were enrolled. Thirty-one patients received bortezomib 1.3 mg/m(2) i.v., on days 1, 4, 8, and 11, for up to six 21-day cycles.

RESULTS

Median age was 63 years (range, 37-82 years). Median number of prior therapies was 2 (range, 1-4). Nine patients had Ann Arbor stage I, 7 patients had stage II, and 16 patients had stage IV. Primary lymphoma localization was the stomach in 14 patients; multiple extranodal sites were present in 10 patients. Among the 29 patients assessable for response, the overall response rate was 48% [95% confidence interval (CI) 29% to 67%], with 9 complete and 5 partial responses. Nine patients experienced stable disease and six had disease progression during therapy. The most relevant adverse events were fatigue, thrombocytopenia, neutropenia, and peripheral neuropathy. After a median follow-up of 24 months, the median duration of response was not reached yet. Five deaths were reported, in two patients due to disease progression.

CONCLUSION

Bortezomib is active in relapsed MALT lymphomas. Further investigations to identify optimal bortezomib dose, schedule, and combination regimens are needed since the frequent detection of dose-limiting peripheral neuropathy.

Flexible modeling improves assessment of prognostic value of C-reactive protein in advanced non-small cell lung cancer

Background:

C-reactive protein (CRP) is gaining credibility as a prognostic factor in different cancers. Cox's proportional hazard (PH) model is usually used to assess prognostic factors. However, this model imposes a priori assumptions, which are rarely tested, that (1) the hazard ratio associated with each prognostic factor remains constant across the follow-up (PH assumption) and (2) the relationship between a continuous predictor and the logarithm of the mortality hazard is linear (linearity assumption).

Methods:

We tested these two assumptions of the Cox's PH model for CRP, using a flexible statistical model, while adjusting for other known prognostic factors, in a cohort of 269 patients newly diagnosed with non-small cell lung cancer (NSCLC).

Results:

In the Cox's PH model, high CRP increased the risk of death (HR=1.11 per each doubling of CRP value, 95% CI: 1.03–1.20, P=0.008). However, both the PH assumption (P=0.033) and the linearity assumption (P=0.015) were rejected for CRP, measured at the initiation of chemotherapy, which kept its prognostic value for approximately 18 months.

Conclusion:

Our analysis shows that flexible modeling provides new insights regarding the value of CRP as a prognostic factor in NSCLC and that Cox's PH model underestimates early risks associated with high CRP.

Variables with time-varying effects and the Cox model: some statistical concepts illustrated with a prognostic factor study in breast cancer

BACKGROUND

The Cox model relies on the proportional hazards (PH) assumption, implying that the factors investigated have a constant impact on the hazard - or risk - over time. We emphasize the importance of this assumption and the misleading conclusions that can be inferred if it is violated; this is particularly essential in the presence of long follow-ups.

METHODS

We illustrate our discussion by analyzing prognostic factors of metastases in 979 women treated for breast cancer with surgery. Age, tumour size and grade, lymph node involvement, peritumoral vascular invasion (PVI), status of hormone receptors (HRec), Her2, and Mib1 were considered.

RESULTS

Median follow-up was 14 years; 264 women developed metastases. The conventional Cox model suggested that all factors but HRec, Her2, and Mib1 status were strong prognostic factors of metastases. Additional tests indicated that the PH assumption was not satisfied for some variables of the model. Tumour grade had a significant time-varying effect, but although its effect diminished over time, it remained strong. Interestingly, while the conventional Cox model did not show any significant effect of the HRec status, tests provided strong evidence that this variable had a non-constant effect over time. Negative HRec status increased the risk of metastases early but became protective thereafter. This reversal of effect may explain non-significant hazard ratios provided by previous conventional Cox analyses in studies with long follow-ups.

CONCLUSIONS

Investigating time-varying effects should be an integral part of Cox survival analyses. Detecting and accounting for time-varying effects provide insights on some specific time patterns, and on valuable biological information that could be missed otherwise.

Reporting of prognostic studies of tumour markers: a review of published articles in relation to REMARK guidelines

Background:

Poor reporting compromises the reliability and clinical value of prognostic tumour marker studies. We review articles to assess the reporting of patients and events using REMARK guidelines, at the time of guideline publication.

Methods:

We sampled 50 prognostic tumour marker studies from higher impact cancer journals between 2006 and 2007. The inclusion criteria were cancer; focus on single biological tumour marker; survival analysis; multivariable analysis; and not gene array or proteomic data. Articles were assessed for the REMARK profile and other REMARK guideline items. We propose a reporting aid, the REMARK profile, motivated by the CONSORT flowchart.

Results:

In 50 studies assessed for the REMARK profile, the number of eligible patients (56% of articles), excluded patients (54%) and patients in analyses (98%) was reported. Only 50% of articles reported the number of outcome events. In multivariable analyses, 54% and 30% of articles reported patient and event numbers for all variables. Of the studies, 66% used archival samples, indicating a potentially biased patient selection. Only 36% of studies reported clearly defined outcomes.

Conclusions:

Good reporting is critical for the interpretability and clinical applicability of prognostic studies. Current reporting of key information, such as the number of outcome events in all patients and subgroups, is poor. Use of the REMARK profile would greatly improve reporting and enhance prognostic research.

Adjuvant platinum-based chemotherapy for early stage cervical cancer

BACKGROUND

Patients with early stage cervical cancer (stages IA2, IB1 or IIA) with risk factors such as lymph node metastasis, lympho vascular space invasion, depth invasion of more than 10mm, microscopic parametrial invasion, non-squamous histology and positive surgical margins have a high risk of recurrence when compared to patients with early stage cervical cancer with no risk factors for recurrence.

OBJECTIVES

To evaluate the effectiveness and safety of platinum-based adjuvant chemotherapy after radical hysterectomy, radiotherapy, or both in the treatment of early stage cervical cancer (stages IA2, IB1 or IIA).

SEARCH STRATEGY

We searched the Cochrane Gynaecological Cancer Group Trials Register, The Cochrane Central Register of Controlled Trials (CENTRAL) on The Cochrane Library Issue 1, 2009), MEDLINE, EMBASE, LILACS, BIOLOGICAL ABSTRACTS and Cancerlit, the National Research Register and Clinical Trials register, with no language restriction. Abstracts of scientific meetings and the citation lists of included studies and other relevant publications were checked through hand searching and experts in the field were contacted to identify further reports of trials.

SELECTION CRITERIA

Randomised controlled trials (RCTs) comparing adjuvant radiotherapy with adjuvant radiotherapy and cisplatin-chemotherapy after radical surgery for early stage cervix cancer were included.

DATA COLLECTION AND ANALYSIS

Two review authors extracted data independently to assess whether the studies met the specified inclusion criteria. Any discrepancies were solved by a third and a forth review author. Meta-analysis was performed using a random effects model, with death and disease progression as outcomes.

MAIN RESULTS

Three trials were included. Two trials enrolling 325 participants, of whom 297 (91%) were assessed and compared radiotherapy and chemotherapy with radiotherapy alone found that adjuvant chemotherapy significantly reduced the risk of death (hazard ratio (HR) = 0.56, 95% confidence interval (CI): 0.36 to 0.87) and disease progression (HR = 0.47, 95%CI: 0.30 to 0.74), with no heterogeneity between trials (I(2) = 0% for both meta-analyses). One trial assessing 71 participants compared chemotherapy followed by radiotherapy with radiotherapy alone and found no significant difference between the two groups (HR = 1.34; 95%CI: 0.24 to 7.66). The median follow up of patients varied from 29 to 42 months.

AUTHORS' CONCLUSIONS

The addition of platinum-based chemotherapy to radiotherapy may offer clinical benefit in the adjuvant treatment of early stage cervical cancer with risk factors for recurrence. However, the evidence is limited because the selected studies were quantitatively and qualitatively limited, with small number of patients and limited period of follow-up.

Adjuvant (post-surgery) chemotherapy for early stage epithelial ovarian cancer

BACKGROUND

Epithelial ovarian cancer is diagnosed in 4500 women in the UK each year of whom 1700 will ultimately die of their disease. Ten to fifteen percent of all cases are diagnosed early when there is still a good possibility for cure. The treatment of early stage disease involves surgery to remove disease often followed by chemotherapy. The largest clinical trials of this adjuvant therapy show an overall survival (OS) advantage with adjuvant platinum based chemotherapy but the precise role of this treatment in sub-groups of patients with differing prognoses needs to be defined.

OBJECTIVES

To systematically review the evidence for adjuvant chemotherapy in early stage epithelial ovarian cancer to determine; firstly whether there is a survival advantage of this treatment over the policy of observation following surgery with chemotherapy reserved for treatment of disease recurrence, and secondly; to determine if clinical sub-groups of differing prognosis based on histological sub-type or completeness of surgical staging, have more or less to gain from chemotherapy following initial surgery.

SEARCH STRATEGY

An electronic search was performed using the Cochrane Gynaecological Cancer Specialised Register, Cochrane Central Register of Controlled Trials (CENTRAL, Issue 2, 2008), MEDLINE (1966 to 2008), EMBASE (1980 to 2008) and CancerLit. The search strategy was developed using free text and medical subject headings (MESH).

SELECTION CRITERIA

The review authors selected those clinical trials that met the inclusion criteria set out based on the populations, interventions, comparisons and outcome measures.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted data and assessed trial quality. Disagreements were resolved by discussion with a third review author. Random effects meta-analyses and sub-group analyses were conducted.

MAIN RESULTS

Five randomised controlled trials (RCTs), enrolling 1277 women, with 46 to 110 months follow-up, met the inclusion criteria. These trials had low risk of bias. Meta-analysis of three trials with adequate data, assessing 1008 women, indicated that women who received adjuvant platinum-based chemotherapy had better overall survival (OS) than those who did not (hazard ratio (HR) 0.71; 95% CI 0.53 to 0.93). Likewise, meta-analysis of four trials with adequate data, assessing 1170 women, indicated that women who received adjuvant chemotherapy had better progression-free survival (PFS) than those who did not (HR 0.67; 95% CI 0.53 to 0.84). The trials included in these meta-analyses gave consistent estimates of the effects of chemotherapy.Sub-group analysis suggested that women who had optimal surgical staging of their disease were unlikely to benefit from adjuvant chemotherapy (HR for OS 1.22; 95% CI 0.63 to 2.37) whereas those who had sub-optimal staging did (HR for OS 0.63; 95% CI 0.46 to 0.85). One trial showed a benefit from adjuvant chemotherapy among women at high risk (HR for OS 0.48; 95% CI 0.32 to 0.72) but not among those at low risk (HR for OS 0.95; 95% CI 0.54 to 1.66). However, these sub-group findings could be due to chance.

AUTHORS' CONCLUSIONS

Adjuvant platinum based chemotherapy is effective in prolonging the survival of the majority of patients who are assessed as having early stage epithelial ovarian cancer. However, even given the limits of sub-group analyses, there is strong evidence that optimal surgical staging identifies patients who have either little or nothing to gain from adjuvant chemotherapy. Taken together with the lack of a survival advantage seen in patients with "low-risk" cancers in the ICON1 trial, it appears safe to withhold adjuvant chemotherapy from optimally staged patients with well differentiated tumours.

Prognostic markers in cancer: the evolution of evidence from single studies to meta-analysis, and beyond

In oncology, prognostic markers are clinical measures used to help elicit an individual patient's risk of a future outcome, such as recurrence of disease after primary treatment. They thus facilitate individual treatment choice and aid in patient counselling. Evidence-based results regarding prognostic markers are therefore very important to both clinicians and their patients. However, there is increasing awareness that prognostic marker studies have been neglected in the drive to improve medical research. Large protocol-driven, prospective studies are the ideal, with appropriate statistical analysis and clear, unbiased reporting of the methods used and the results obtained. Unfortunately, published prognostic studies rarely meet such standards, and systematic reviews and meta-analyses are often only able to draw attention to the paucity of good-quality evidence. We discuss how better-quality prognostic marker evidence can evolve over time from initial exploratory studies, to large protocol-driven primary studies, and then to meta-analysis or even beyond, to large prospectively planned pooled analyses and to the initiation of tumour banks. We highlight articles that facilitate each stage of this process, and that promote current guidelines aimed at improving the design, analysis, and reporting of prognostic marker research. We also outline why collaborative, multi-centre, and multi-disciplinary teams should be an essential part of future studies.

[Strengthening the reporting of observational studies in epidemiology (STROBE): explanation and elaboration]

Much medical research is observational. The reporting of observational studies is often of insufficient quality. Poor reporting hampers the assessment of the strengths and weaknesses of a study and the generalisability of its results. Taking into account empirical evidence and theoretical considerations, a group of methodologists, researchers, and editors developed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) recommendations to improve the quality of reporting of observational studies. The STROBE Statement consists of a checklist of 22 items, which relate to the title, abstract, introduction, methods, results and discussion sections of articles. Eighteen items are common to cohort studies, case-control studies and cross-sectional studies and four are specific to each of the three study designs. The STROBE Statement provides guidance to authors about how to improve the reporting of observational studies and facilitates critical appraisal and interpretation of studies by reviewers, journal editors and readers. This explanatory and elaboration document is intended to enhance the use, understanding, and dissemination of the STROBE Statement. The meaning and rationale for each checklist item are presented. For each item, one or several published examples and, where possible, references to relevant empirical studies and methodological literature are provided. Examples of useful flow diagrams are also included. The STROBE Statement, this document, and the associated Web site (http://www.strobe-statement.org/) should be helpful resources to improve reporting of observational research.

Long-term outcome following Helicobacter pylori eradication in a retrospective study of 105 patients with localized gastric marginal zone B-cell lymphoma of MALT type

BACKGROUND

Treatment aimed at eradicating Helicobacter pylori infection results in lymphoma remission in most localized gastric mucosa-associated lymphoid tissue (MALT) lymphomas. The aim of this survey is to investigate the long-term effect of this therapeutic approach in a large series of patients.

METHODS

One hundred and five patients with localized gastric MALT lymphoma were initially treated only with H. pylori eradication regimens. Lymphoma responses were graded using the Wotherspoon score.

RESULTS

Helicobacter pylori, detected by histology in 81% of cases, was eradicated in all positive patients. Histological regression of the lymphoma was achieved in 78 of 102 assessable patients [76%, 95% confidence interval (CI): 67% to 84%] with complete remission (score 0-2) in 66 and partial remission (score 3) in 12. At a median follow-up time of 6.3 years, histological remission was consistently confirmed in 33 of 74 assessable patients, while 25 had score fluctuations (from 0 to 4) and 13 presented a lymphoma relapse (score 5). Only one patient had a distant progression. Transformation to a large-cell lymphoma was seen in two cases. The 5- and 10-year overall survival is 92% (95% CI: 84% to 96%) and 83% (95% CI: 70% to 91%), respectively. Only one patient died of lymphoma after transformation to a high-grade lymphoma.

CONCLUSIONS

Helicobacter pylori eradication resulted in complete lymphoma remission in the majority of cases. Long-term clinical disease control was achieved in most patients. A watch and wait policy appears to be safe in patients with minimal residual disease or histological-only local relapse.

Adjuvant (post-surgery) chemotherapy for early stage epithelial ovarian cancer

BACKGROUND

Epithelial ovarian cancer kills about 1700 in the UK each year. Ten to fifteen percent of all cases are diagnosed early when there is still a good chance of cure. The treatment of early stage disease involves surgery to remove disease often followed by chemotherapy. The largest clinical trials of this adjuvant therapy show an overall survival (OS) advantage with adjuvant platinum based chemotherapy but the precise role of this treatment in sub-groups of patients with differing prognoses needs to be defined.

OBJECTIVES

To systematically review the evidence for adjuvant chemotherapy in early stage epithelial ovarian cancer to determine; firstly whether there is a survival advantage of this treatment over the policy of observation following surgery with chemotherapy reserved for treatment of disease recurrence, and secondly; to determine if clinical sub-groups of differing prognosis based on histological sub-type or completeness of surgical staging, have more or less to gain from chemotherapy following initial surgery.

SEARCH STRATEGY

An electronic search was performed using the Cochrane Gynaecological Cancer Specialised Register, Cochrane Central Register of Controlled Trials (CENTRAL, Issue 2 2008), MEDLINE (1966 to 2008), EMBASE (1980 to 2008) and CancerLit. The search strategy was developed using free text and medical subject headings (MESH). This yielded a large number of article titles which were sifted down by two review authors to a limited number of articles, the full text versions of which were independently reviewed to select out clinical trials of direct and specific relevance to the review question. Hand searches of the clinical literature were conducted where appropriate to identify additional full-text papers or abstracts of other directly relevant clinical trials.

SELECTION CRITERIA

The review authors selected those clinical trials that met the inclusion criteria set out based on the populations, interventions , comparisons and outcome measures as detailed in the full text review.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted data and assessed trial quality. Disagreements were resolved by discussion or a third reviewer. Random effects meta-analyses and sub-group analyses were conducted.

MAIN RESULTS

Five randomised controlled trials (RCTs), enrolling 1277 women, with 46 to 110 months follow-up, met our inclusion criteria. These trials had low risk of bias. Meta-analysis of three trials with adequate data, assessing 1008 women, indicated that women who received adjuvant platinum-based chemotherapy had better overall survival (OS) than those who did not (hazard ratio (HR) 0.71; 95% CI 0.53 to 0.93). Likewise, meta-analysis of four trials with adequate data, assessing 1170 women, indicated that women who received adjuvant chemotherapy had better progression-free survival (PFS) than those who did not (HR 0.67; 95% CI 0.53 to 0.84). The trials included in these meta-analyses gave consistent estimates of the effects of chemotherapy.Sub-group analysis suggested that women who had optimal surgical staging of their disease were unlikely to benefit from adjuvant chemotherapy (HR for OS 1.22; 95% CI 0.63 to 2.37) whereas those who had sub-optimal staging did (HR for OS 0.63; 95% CI 0.46 to 0.85). One trial showed a benefit from adjuvant chemotherapy among women at high risk (HR for OS 0.48; 95% CI 0.32 to 0.72) but not among those at low risk (HR for OS 0.95; 95% CI 0.54 to 1.66). However, these sub-group findings could be due to chance.

AUTHORS' CONCLUSIONS

Adjuvant platinum based chemotherapy is effective in prolonging the survival of the majority of patients who are assessed as having early stage epithelial ovarian cancer. However, even given the limits of sub-group analyses, there is strong evidence that optimal surgical staging identifies patients who have either little or nothing to gain from adjuvant chemotherapy. Taken together with the lack of a survival advantage seen in patients with "low-risk" cancers in the ICON1 trial, it appears safe to withhold adjuvant chemotherapy from optimally staged patients with well differentiated tumours.

Accelerated failure time models provide a useful statistical framework for aging research

Survivorship experiments play a central role in aging research and are performed to evaluate whether interventions alter the rate of aging and increase lifespan. The accelerated failure time (AFT) model is seldom used to analyze survivorship data, but offers a potentially useful statistical approach that is based upon the survival curve rather than the hazard function. In this study, AFT models were used to analyze data from 16 survivorship experiments that evaluated the effects of one or more genetic manipulations on mouse lifespan. Most genetic manipulations were found to have a multiplicative effect on survivorship that is independent of age and well-characterized by the AFT model "deceleration factor". AFT model deceleration factors also provided a more intuitive measure of treatment effect than the hazard ratio, and were robust to departures from modeling assumptions. Age-dependent treatment effects, when present, were investigated using quantile regression modeling. These results provide an informative and quantitative summary of survivorship data associated with currently known long-lived mouse models. In addition, from the standpoint of aging research, these statistical approaches have appealing properties and provide valuable tools for the analysis of survivorship data.

Survival End Point Reporting in Randomized Cancer Clinical Trials: A Review of Major Journals

Purpose

Several publications showed that the standards for reporting randomized clinical trials (RCTs) might not be entirely suitable. Our aim was to evaluate the reporting of survival end points in cancer RCTs.

Methods

A search in MEDLINE databases identified 274 cancer RCTs published in 2004 in four general medical journals and four clinical oncology journals. Eligible articles were those that reported primary analyses of RCT with survival end points. Methodologists reviewed and scored the articles according to seven key points: prevalence of complete definition of survival end points (time of origin, survival events, censoring events) and relevant information about their analyses (estimation or effect size, precision, number of events, patients at risk). Concordance of key points was evaluated from a random subsample.

Results

After screening, 125 articles were selected; 104 trials were phase III (83%) and 98 publications (78%) were obtained from oncology journals. Among these RCTs, a total of 267 survival end points were recorded, and overall survival (OS) was the most frequent outcome (118 terms, 44%). Survival terms were totally defined for 113 end points (42%) in 65 articles (52%). Accurate information about analysis was retrieved for 73 end points (27%) in 40 articles (32%). The less well-defined information was the number of patients at risk (55%). The reliability was good (κ = 0.72). Finally, according to the key points, optimal reporting was found in 33 end points (12%) or 10 publications.

Conclusion

A majority of articles failed to provide a complete reporting of survival end points, thus adding another source of uncontrolled variability.

Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration

Much medical research is observational. The reporting of observational studies is often of insufficient quality. Poor reporting hampers the assessment of the strengths and weaknesses of a study and the generalisability of its results. Taking into account empirical evidence and theoretical considerations, a group of methodologists, researchers, and editors developed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) recommendations to improve the quality of reporting of observational studies. The STROBE Statement consists of a checklist of 22 items, which relate to the title, abstract, introduction, methods, results and discussion sections of articles. Eighteen items are common to cohort studies, case-control studies and cross-sectional studies and four are specific to each of the three study designs. The STROBE Statement provides guidance to authors about how to improve the reporting of observational studies and facilitates critical appraisal and interpretation of studies by reviewers, journal editors and readers. This explanatory and elaboration document is intended to enhance the use, understanding, and dissemination of the STROBE Statement. The meaning and rationale for each checklist item are presented. For each item, one or several published examples and, where possible, references to relevant empirical studies and methodological literature are provided. Examples of useful flow diagrams are also included. The STROBE Statement, this document, and the associated Web site (http://www.strobe-statement.org/) should be helpful resources to improve reporting of observational research.

Quantification of outpatient management and hospitalization of patients with high-risk myelodysplastic syndrome treated with low-dose decitabine

Intravenous low-dose 5-aza-2'-deoxycytidine (decitabine) in patients with advanced myelodysplastic syndrome (MDS) yields an approximately 50% overall response rate, including 20-25% complete remission. Decitabine-treated MDS patients can be managed as outpatients after completion of a 3-day infusion schedule. In-hospital nights (IHNs), overall survival (OS), and remaining life spent in hospital were evaluated and compared to a matched control group receiving different standard treatments. Between July 1992 and September 2001, 99 high-risk MDS patients, median age 70 years (range 49-86), were treated with low-dose decitabine. Durations of all hospitalizations were recorded. For matched-pair analysis, 44 decitabine-treated patients were matched to 44 MDS patients according to International Prognostic Scoring System classification, period of diagnosis, age, French-American-British classification, and gender. Median number of IHN across all patients was 56 and survival was 481 days, resulting in 84% of remaining life spent at home. In the matched-pair analysis, the median number of IHN was 57 in the decitabine group vs. 50 in the control group. Median survival was 400 vs. 371 days for the decitabine and control groups, respectively. Median number of remaining life spent at home was identical (83% for both groups). Matched patients who received only best supportive care (n=12) had a shorter median survival than the decitabine patients (234 vs. 400 days), and the proportion of remaining life spent at home was slightly greater (82 vs. 77%). Interestingly, matched patients with induction therapy showed comparable IHN, OS, and remaining life spent at home. In conclusion, high-risk MDS patients treated with low-dose decitabine have better survival, and spend comparable time in hospital than patients treated with supportive treatment.

The in- or exclusion of non-breast cancer related death and contralateral breast cancer significantly affects estimated outcome probability in early breast cancer

A wide variation of definitions of recurrent disease and survival are used in the analyses of outcome of patients with early breast cancer. Explicit definitions with details both on endpoints and censoring are provided in less than half of published studies. We evaluated the effects of various definitions of survival and recurrent disease on estimated outcome in a prospectively determined cohort of 463 patients with primary breast cancer. Outcome estimates were determined both by the Kaplan-Meier and a competing risk method. In- or exclusion of contralateral breast cancer or non-disease related death in the definition of recurrent disease or survival significantly affects estimated outcome probability. The magnitude of this finding was dependent on patient-, tumour-, and treatment characteristics. Knowledge of the contribution of non-disease related death or contralateral breast cancer to estimated recurrent disease rate and overall death rate is indispensable for a correct interpretation and comparison of outcome analyses.

Meta-analyses of chemotherapy for locally advanced and metastatic pancreatic cancer

PURPOSE

There are a large number of randomized controlled trials involving chemotherapy in the management of advanced pancreatic cancer. Several chemotherapeutic agents, either alone or in combination with other chemotherapy or novel agents, have been used. The aim of these meta-analyses was to examine the different therapeutic approaches, and the comparisons examined were as follows: chemotherapy versus best supportive care; fluorouracil (FU) versus FU combination chemotherapy; gemcitabine versus FU; and gemcitabine versus gemcitabine combination chemotherapy.

METHODS

Relevant trials were identified by searching databases, trial registers, and conference proceedings. The primary end point was overall survival.

RESULTS

One hundred thirteen randomized controlled trials were identified, of which 51 trials involving 9,970 patients met the inclusion criteria. Chemotherapy improved survival compared with best supportive care (hazard ratio [HR] = 0.64; 95% CI, 0.42 to 0.98). FU-based combination chemotherapy did not result in better overall survival compared with FU alone (HR = 0.94; 95% CI, 0.82 to 1.08). There was insufficient evidence of a survival difference between gemcitabine and FU, but the wide CI includes clinically important differences in both directions, making a clear conclusion difficult (HR = 0.75; 95% CI, 0.42 to 1.31). Survival was improved after gemcitabine combination chemotherapy compared with gemcitabine alone (HR = 0.91; 95% CI, 0.85 to 0.97).

CONCLUSION

There was a significant survival benefit for chemotherapy over best supportive care and gemcitabine combinations over gemcitabine alone. This supports the use of gemcitabine-based combination chemotherapy in the treatment of advanced pancreatic cancer.

Tolerability of accelerated chest irradiation and impact on survival of prophylactic cranial irradiation in patients with limited-stage small cell lung cancer: review of a single institution's experience

INTRODUCTION

Evidence that has been published in the last decade indicates that in patients with limited-stage small-cell lung cancer (SCLC), hyperfractionated accelerated thoracic radiotherapy (RT) given twice daily and prophylactic cranial irradiation (PCI) have each separately improved survival. Concerns about the toxicities associated with these treatments and uncertainty about their impact on survival outside the trial setting may have restricted the extent to which they have been incorporated into standard treatment protocols. We have reviewed the experience at Peter MacCallum Cancer Centre to determine the tolerability of these treatments in routine practice and to determine their effects on survival.

METHODS

A retrospective review of patients with limited-stage SCLC receiving a radical course of thoracic RT between June 1998 and May 2002, including either conventional fractionation at 50 Gy for 5 weeks, or hyperfractionated accelerated RT at 45 Gy for 3 weeks. Patients achieving a complete response were offered PCI at 36 Gy in 18 fractions. The main outcomes recorded were RT toxicity (graded using CTCAE v. 3.0 and RTOG/EORTC late scoring criteria), response, relapse-free survival, and overall survival.

RESULTS

Ninety patients were identified as having undergone radical-intent thoracic RT, with a median potential follow-up of 4.2 years. Fifty-seven patients (63%) were treated with hyperfractionated accelerated RT, and 33 (37%) were treated with conventional fractionation. Forty-six patients (51%) received PCI. Patients receiving hyperfractionated accelerated RT compared with conventional fractionation had higher rates of grade 3 and 4 esophagitis (14% versus 6%; p = 0.312), a higher rate of treatment interruptions (12% versus 3%; p = 0.250), and a higher hospital admission rate (39% versus 15%; p = 0.031). The majority of patients were able to complete the planned treatment, and there were no treatment-related deaths. Median survival for all patients from commencement of RT was 14.2 months (95% confidence interval [CI]: 11.9-18.1 months), and survival at 2 years was 24.8% (95% CI: 16.9-35.0%). On multifactor analysis, the only factor associated with longer survival was PCI (hazard ratio = 0.40; p < 0.001).

CONCLUSIONS

Hyperfractionated accelerated RT was more toxic than conventional fractionation, but it was possible to deliver treatment as planned in the majority of patients. PCI was associated with improved survival. Both treatments can be incorporated into routine practice.

Practical methods for incorporating summary time-to-event data into meta-analysis

BACKGROUND

In systematic reviews and meta-analyses, time-to-event outcomes are most appropriately analysed using hazard ratios (HRs). In the absence of individual patient data (IPD), methods are available to obtain HRs and/or associated statistics by carefully manipulating published or other summary data. Awareness and adoption of these methods is somewhat limited, perhaps because they are published in the statistical literature using statistical notation.

METHODS

This paper aims to 'translate' the methods for estimating a HR and associated statistics from published time-to-event-analyses into less statistical and more practical guidance and provide a corresponding, easy-to-use calculations spreadsheet, to facilitate the computational aspects.

RESULTS

A wider audience should be able to understand published time-to-event data in individual trial reports and use it more appropriately in meta-analysis. When faced with particular circumstances, readers can refer to the relevant sections of the paper. The spreadsheet can be used to assist them in carrying out the calculations.

CONCLUSION

The methods cannot circumvent the potential biases associated with relying on published data for systematic reviews and meta-analysis. However, this practical guide should improve the quality of the analysis and subsequent interpretation of systematic reviews and meta-analyses that include time-to-event outcomes.

Experience with cardiac valve operations in Cape York Peninsula and the Torres Strait Islands, Australia

OBJECTIVE

To describe the outcome of valve surgery, for rheumatic heart disease (RHD) and non-RHD, in residents of Cape York Peninsula and the Torres Strait Islands referred to the Cairns Base Hospital specialist outreach service.

DESIGN AND PARTICIPANTS

Retrospective review of medical records on all patients residing in the outreach area who had surgery for valvular heart disease between 1 January 1992 and 31 December 2004.

MAIN OUTCOME MEASURES

Operation type and perioperative characteristics; 5- and 10-year survival rates; reoperation rates; complications.

RESULTS

Forty-seven patients met the selection criteria; the median age was 40 years (range, 4-76 years); and 39 patients were Indigenous. RHD was the predominant cause of valve dysfunction (30/47 patients). Thirty-seven patients had valve replacements, six had valve repair and four had balloon valvotomy as the initial procedure. There were three bleeding complications, two episodes of operated valve endocarditis, and six embolic complications. There were nine valve-related deaths (six in the first 5 years). At 5 years, all seven patients who had had valve repair or balloon valvotomy were alive. Seven of the 47 patients required reoperation. Survival analysis showed freedom from valve-related deaths to be 83% (95% CI, 66%-92%) at 5 years and 61% (95% CI, 33%-80%) at 10 years. Freedom from reoperation at 5 years was 88% (95% CI, 71%-95%). Among the 30 patients with RHD, freedom from valve-related death was 80% (95% CI, 60%-92%) at 5 years and 52% (95% CI, 21%-75%) at 10 years. In patients with RHD, freedom from reoperation at 5 years was 87% (95% CI, 65%-96%).

CONCLUSION

Valvular heart disease results in substantial morbidity and mortality, despite intervention. Efforts need to focus on prevention of rheumatic fever and closer follow-up.

Malignant ureteral obstruction: outcomes after intervention. Have things changed?

PURPOSE

We reviewed the clinical outcome for patients who underwent decompression of malignant ureteral obstruction by analyzing potential prognostic factors, technical success, complication rates and days spent in hospital.

MATERIALS AND METHODS

Retrospective analysis of 102 patients who underwent decompression for malignant ureteral obstruction from 1991 to 2003 was performed. Data on overall survival, prognostic factors, technical failure, complications and days of hospitalization after decompression were examined.

RESULTS

Median overall survival was 6.8 months (95% CI 4.8-9.3) and the overall survival rate at 12 months was 29% (95% CI 21%-39%). Univariate analyses found inferior overall survival associated with patients having metastases (p=0.041), undergoing nephrostomy (p=0.046), prior treatment for cancer (p=0.024) and diagnosis of malignant ureteral obstruction in previously established malignancy (p=0.043). After multivariate analysis poor prognostic factors were presence of metastases (p=0.020) and diagnosis of malignant ureteral obstruction in previously established malignancy (p=0.039). Unfavorable cohorts with 3 or 4 unfavorable baseline risk factors had an inferior overall survival (p=0.008) with 12-month overall survival rates of 12% to 19%. Initial decompression of malignant ureteral obstruction failed in 6% of patients (95% CI 2%-12%) and complications were experienced by 53% (95% CI 43%-63%). Patients were more likely to experience a complication if they had therapy after decompression (p=0.03). The median percentage of their remaining lifetime spent in hospital was 17.4% (range 0.21% to 100%).

CONCLUSIONS

The overall survival of patients with malignant ureteral obstruction remains poor. Prognostic factors for decreased overall survival and prolonged hospital stay have been identified. Although the technical success of decompression has improved the subsequent complication rate is still high.

Systematic review, including meta-analyses, on the management of locally advanced pancreatic cancer using radiation/combined modality therapy

There is no consensus on the management of locally advanced pancreatic cancer, with either chemotherapy or combined modality approaches being employed (Maheshwari and Moser, 2005). No published meta-analysis (Fung et al, 2003; Banu et al, 2005; Liang, 2005; Bria et al, 2006; Milella et al, 2006) has included randomised controlled trials employing radiation therapy. The aim of this systematic review was to compare the following: (i) chemoradiation followed by chemotherapy (combined modality therapy) vs best supportive care (ii) radiotherapy vs chemoradiation (iii) radiotherapy vs combined modality therapy (iv) chemotherapy vs combined modality therapy (v) 5FU-based combined modality treatment vs another-agent-based combined modality therapy. Relevant randomised controlled trials were identified by searching databases, trial registers and conference proceedings. The primary end point was overall survival and secondary end points were progression-free survival/time-to-progression, response rate and adverse events. Survival data were summarised using hazard ratio (HR) and response-rate/adverse-event data with relative risk. Eleven trials involving 794 patients met the inclusion criteria. Length of survival with chemoradiation was increased compared with radiotherapy alone (two trials, 168 patients, HR 0.69; 95% confidence interval (CI) 0.51-0.94), but chemoradiation followed by chemotherapy did not lead to a survival advantage over chemotherapy alone (two trials, 134 patients, HR 0.79; CI 0.32-1.95). Meta-analyses could not be performed for the other comparisons. A survival benefit was demonstrated for chemoradiation over radiotherapy alone. Chemoradiation followed by chemotherapy did not demonstrate any survival advantage over chemotherapy alone, but important clinical differences cannot be ruled out due to the wide CI.

Gestational diabetes mellitus: clinical predictors and long-term risk of developing type 2 diabetes: a retrospective cohort study using survival analysis

OBJECTIVE

We sought to determine the long-term risk of type 2 diabetes following a pregnancy complicated by gestational diabetes mellitus (GDM) and assess what maternal antepartum, postpartum, and neonatal factors are predictive of later development of type 2 diabetes.

RESEARCH DESIGN AND METHODS

This was a retrospective cohort study using survival analysis on 5,470 GDM patients and 783 control subjects who presented for postnatal follow-up at the Mercy Hospital for Women between 1971 and 2003.

RESULTS

Risk of developing diabetes increased with time of follow-up for both groups and was 9.6 times greater for patients with GDM. The cumulative risk of developing type 2 diabetes for the GDM patients was 25.8% at 15 years postdiagnosis. Predictive factors for the development of type 2 diabetes were use of insulin (hazard ratio 3.5), Asian origin compared with Caucasian (2.1), and 1-h blood glucose (1.3 for every 1 mmol increase above 10.1 mmol). BMI was associated with an increased risk of developing type 2 diabetes but did not meet the assumption of proportional hazards required for valid inference when using Cox proportional hazards.

CONCLUSIONS

While specific predictive factors for the later development of type 2 diabetes can be identified in the index pregnancy, women with a history of GDM, as a group, are worthy of long-term follow-up to ameliorate their excess cardiovascular risk.

Radiotherapy for low-grade gastric marginal zone lymphoma: a retrospective study

BACKGROUND

We evaluated the efficacy and toxicity of radiotherapy (RT) in patients with low-grade gastric marginal zone lymphoma.

METHODS

A retrospective review of consecutive cases of gastric marginal zone lymphoma treated by radical RT at the Peter MacCallum Cancer Centre and Radiation Oncology Victoria between January 1980 and September 2003 was carried out.

RESULTS

Eighteen patients (11 men and 7 women) were identified. The median age at commencement of RT was 65 years (range 42-84 years). Prior treatment included Helicobacter pylori eradication in 12 patients, chemotherapy in 7 and surgery in 2, whereas 2 patients had no prior therapy. The median time to progression after commencement of last treatment before RT was 4.8 months (range 0-129.4 months). The radiation fields included the stomach plus perigastric and coeliac nodes in 15 patients (83%), stomach plus spleen in 2 patients (11%) and stomach plus para-aortic nodes in 1 patient (6%). The median RT dose was 30 Gy (range 30-36 Gy) in a median 20 fractions (range 17-24 fractions). One patient required treatment interruption for acute toxicity. A complete response on post-RT biopsies was achieved in 17 of 18 patients (94%). With a median follow up of 4.5 years after RT, 3 of these 17 patients (18%) have had a recurrence. At the last follow up, 11 patients were alive in continuous complete histological remission. No late renal toxicity was identified.

CONCLUSION

Radiotherapy is an effective, well-tolerated treatment for patients with low-grade gastric marginal zone lymphoma, including those who have had prior therapy.

REporting recommendations for tumor MARKer prognostic studies (REMARK)

Despite years of research and hundreds of reports on tumor markers in oncology, the number of markers that have emerged as clinically useful is pitifully small. Often initially reported studies of a marker show great promise, but subsequent studies on the same or related markers yield inconsistent conclusions or stand in direct contradiction to the promising results. It is imperative that we attempt to understand the reasons that multiple studies of the same marker lead to differing conclusions. A variety of methodologic problems have been cited to explain these discrepancies. Unfortunately, many tumor marker studies have not been reported in a rigorous fashion, and published articles often lack sufficient information to allow adequate assessment of the quality of the study or the generalizability of study results. The development of guidelines for the reporting of tumor marker studies was a major recommendation of the National Cancer Institute-European Organisation for Research and Treatment of Cancer (NCI-EORTC) First International Meeting on Cancer Diagnostics in 2000. As for the successful CONSORT initiative for randomized trials and for the STARD statement for diagnostic studies, we suggest guidelines to provide relevant information about the study design, pre-planned hypotheses, patient and specimen characteristics, assay methods, and statistical analysis methods. In addition, the guidelines suggest helpful presentations of data and important elements to include in discussions. The goal of these guidelines is to encourage transparent and complete reporting so that the relevant information will be available to others to help them to judge the usefulness of the data and understand the context in which the conclusions apply.

Primer: an evidence-based approach to prognostic markers

Prognostic markers can help to identify patients at different degrees of risk for specific outcomes, facilitate treatment choice, and aid patient counseling. Compared with other research designs, prognostic studies have been relatively neglected in the broad efforts to improve the quality of medical research, despite their ubiquity. Large protocol-driven, prospective studies are the ideal, with clear, unbiased reporting of the methods used and the results obtained. Unfortunately, published prognostic studies rarely meet such standards, and in this article we discuss their main problems and how they can be improved. In particular, an evidence-based approach to prognostic markers is required, as it is usually difficult to ascertain the benefit of a marker from single studies and a clear view is only likely to emerge from looking across multiple studies. Current systematic reviews and meta-analyses often fail to provide clear evidence-based answers, and rather only draw attention to the paucity of good-quality evidence. Prospectively planned pooled analyses of high-quality studies, along with general availability of individual patient data and adherence to reporting guidelines, would help alleviate many of these problems.

Fludarabine phosphate-CVP in patients over 60 years of age with advanced, low-grade and follicular lymphoma: a dose-finding study

The aim of this study was to establish a safe and effective regimen of fludarabine phosphate, cyclophosphamide, vincristine and prednisone (F-CVP) as first-line treatment for elderly patients with advanced, low-grade non-Hodgkin's lymphoma. Twenty-three patients >60 years were assigned successively to eight treatment cycles (Dose level 1: low F, low CV [n=4]; 2A: high F, low CV [n=8]; 2B: low F, high CV [n=4]; 3: high F, high CV [n=7]). High and low levels were: F, 25 and 20mg/m(2), respectively (Days 1-5); C, 750 and 500 mg/m(2), respectively (Day 1); and V, 1.4 and 1mg/m(2), respectively (Day 1). Patients received P at 40 mg/m(2) on Days 1-5. Response was assessed after Cycles 2, 4, 6 and 8. At level 3, dose-limiting toxicity (opportunistic infections and neutropenia) became evident, particularly after Cycle 6. Further patients were recruited at Dose level 2A. All regimens proved effective, with an OR rate of 78% (65% CR), and 3-year survival of 65% (+/-10%). Among 18 responders, 51% were still in response at 3 and 5 years. The study shows that this combination therapy is highly effective. The addition of F to CVP at Dose level 2A was feasible and increased the CR rate, with good tolerability in elderly patients.

REporting recommendations for tumour MARKer prognostic studies (REMARK)

Despite years of research and hundreds of reports on tumour markers in oncology, the number of markers that have emerged as clinically useful is pitifully small. Often initially reported studies of a marker show great promise, but subsequent studies on the same or related markers yield inconsistent conclusions or stand in direct contradiction to the promising results. It is imperative that we attempt to understand the reasons that multiple studies of the same marker lead to differing conclusions. A variety of methodologic problems have been cited to explain these discrepancies. Unfortunately, many tumour marker studies have not been reported in a rigorous fashion, and published articles often lack sufficient information to allow adequate assessment of the quality of the study or the generalisability of study results. The development of guidelines for the reporting of tumour marker studies was a major recommendation of the National Cancer Institute-European Organisation for Research and Treatment of Cancer (NCI-EORTC) First International Meeting on Cancer Diagnostics in 2000. As for the successful CONSORT initiative for randomised trials and for the STARD statement for diagnostic studies, we suggest guidelines to provide relevant information about the study design, pre-planned hypotheses, patient and specimen characteristics, assay methods, and statistical analysis methods. In addition, the guidelines suggest helpful presentations of data and important elements to include in discussions. The goal of these guidelines is to encourage transparent and complete reporting so that the relevant information will be available to others to help them to judge the usefulness of the data and understand the context in which the conclusions apply.

An overview of methods and empirical comparison of aggregate data and individual patient data results for investigating heterogeneity in meta-analysis of time-to-event outcomes

Combining the results of individual studies using meta-analysis may be undertaken using either aggregate data (AD) or individual patient data (IPD). In any meta-analysis it is important to consider statistical heterogeneity between studies. Potential sources of heterogeneity can be explored using regression models with either AD or IPD. An overview of approaches and empirical assessment of how the results and conclusions differ from these analyses is undertaken using a meta-analysis of five randomized controlled trials comparing two antiepileptic drugs with time-to-event outcomes. Alternative meta-regression models using AD are compared to stratified Cox regression models using IPD. Age as a potential cause of heterogeneity is detected by both AD and IPD regression models. Time from first ever seizure to randomization is only identified by some AD models. A more thorough explanation of heterogeneity is obtained from the model using IPD but further empirical evidence comparing IPD and AD results are needed.

REporting recommendations for tumour MARKer prognostic studies (REMARK)

Despite years of research and hundreds of reports on tumour markers in oncology, the number of markers that have emerged as clinically useful is pitifully small. Often initially reported studies of a marker show great promise, but subsequent studies on the same or related markers yield inconsistent conclusions or stand in direct contradiction to the promising results. It is imperative that we attempt to understand the reasons that multiple studies of the same marker lead to differing conclusions. A variety of methodological problems have been cited to explain these discrepancies. Unfortunately, many tumour marker studies have not been reported in a rigorous fashion, and published articles often lack sufficient information to allow adequate assessment of the quality of the study or the generalisability of the study results. The development of guidelines for the reporting of tumour marker studies was a major recommendation of the US National Cancer Institute and the European Organisation for Research and Treatment of Cancer (NCI-EORTC) First International Meeting on Cancer Diagnostics in 2000. Similar to the successful CONSORT initiative for randomised trials and the STARD statement for diagnostic studies, we suggest guidelines to provide relevant information about the study design, preplanned hypotheses, patient and specimen characteristics, assay methods, and statistical analysis methods. In addition, the guidelines suggest helpful presentations of data and important elements to include in discussions. The goal of these guidelines is to encourage transparent and complete reporting so that the relevant information will be available to others to help them to judge the usefulness of the data and understand the context in which the conclusions apply.

Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK)

Despite years of research and hundreds of reports on tumor markers in oncology, the number of markers that have emerged as clinically useful is pitifully small. Often, initially reported studies of a marker show great promise, but subsequent studies on the same or related markers yield inconsistent conclusions or stand in direct contradiction to the promising results. It is imperative that we attempt to understand the reasons that multiple studies of the same marker lead to differing conclusions. A variety of methodologic problems have been cited to explain these discrepancies. Unfortunately, many tumor marker studies have not been reported in a rigorous fashion, and published articles often lack sufficient information to allow adequate assessment of the quality of the study or the generalizability of study results. The development of guidelines for the reporting of tumor marker studies was a major recommendation of the National Cancer Institute-European Organisation for Research and Treatment of Cancer (NCI-EORTC) First International Meeting on Cancer Diagnostics in 2000. As for the successful CONSORT initiative for randomized trials and for the STARD statement for diagnostic studies, we suggest guidelines to provide relevant information about the study design, preplanned hypotheses, patient and specimen characteristics, assay methods, and statistical analysis methods. In addition, the guidelines suggest helpful presentations of data and important elements to include in discussions. The goal of these guidelines is to encourage transparent and complete reporting so that the relevant information will be available to others to help them to judge the usefulness of the data and understand the context in which the conclusions apply.

Meta-analysis when only the median survival times are known: a comparison with individual patient data results

BACKGROUND

The hazard ratio (HR) is the most appropriate measure for time to event outcomes such as survival. In systematic reviews, HRs can be calculated either from the raw trial data obtained as part of an individual patient data (IPD) meta-analysis or from the appropriate trial-level summary statistics. However, the information required for the latter are seldom reported in sufficient detail to allow reviewers to calculate HRs. In contrast, the median survival and survival rates at specific time points are frequently presented. We aimed to evaluate retrospectively the performance of meta-analyses using median survival times and survival rates by comparing them with meta-analyses using IPD to calculate HRs.

METHODS

IPD from thirteen published meta-analyses (MAs) in cancers with high mortality rates were used. Median survival and survival rates were calculated from the IPD rather than taken from publications so that the same trials, patients, and extended follow-up are used in each analysis.

RESULTS AND CONCLUSIONS

We show that using median survival times or survival rates at a particular point in time are not reasonable surrogate measures for meta-analyses of survival outcomes and that, wherever possible, HRs should be calculated. Individual trial publications reporting on time to event outcomes, therefore, should provide more detailed statistical information, preferably logHRs and their variances, or their estimators.

Circulating biomarkers from tumour bulk to tumour machinery: promises and pitfalls

In this paper, we provide a working classification for circulating biomarkers according to their potential clinical application. We broadly divided biomarkers into four groups: (i) biomarkers of cancer risk, (ii) biomarkers of tumour-host interactions, (iii) biomarker of tumour burden, and (iv) function-related biomarkers. We hope this classification will provide a framework to which the results of future studies can be added. We also discuss the promises and pitfalls in the optional use of biomarkers in oncology.

Statistical Time to Event Analysis in the Social Sciences: Modeling Hazard Rate and Duration in Finance

Abstract. The relevance of statistical time to event analysis in the social sciences has proved to be of great importance in the last few years, especially in applications related to labor-market analysis, employment and/or unemployment issues, duration of strikes, and survival of new firms, and in financial applications related to the time a company spends in a given status, for example, bankruptcy. We review some of the techniques that have proved to be adequate for analyzing this type of data and the conditions they require for their proper use. In addition, we extend these techniques in order to be able to analyze specific and more complex situations by using a more general and flexible model. All of these techniques and their extensions are illustrated with an example that studies the duration of firms under bankruptcy in the United States.

Sensitivity analyses allowed more appropriate and reliable meta-analysis conclusions for multiple outcomes when missing data was present

OBJECTIVE

A major problem for meta-analysis of multiple outcomes is the unavailability of some estimates from published and unpublished studies. Dissemination bias, in how and what outcomes are reported or published, may be causing this incompleteness. This article illustrates these problems and presents possible sensitivity analyses to allow the most reliable conclusions.

STUDY DESIGN AND SETTING

In a systematic review of prognostic marker MYC-N in neuroblastoma, meta-analysis for overall survival (OS) and disease-free survival (DFS) was of interest. Only 17 published studies enabled extraction of both outcome estimates, 25 enabled only DFS, 39 enabled only OS, and 70 enabled neither outcome. Unidentified unpublished studies may also exist. We assessed the robustness of the pooled estimates to the problem of missing information. Because OS and DFS estimates seemed to be related, we used the known outcome estimates to predict estimates known to be missing, and combined this approach with existing methods for assessing dissemination bias.

RESULTS

The results of the sensitivity analyses suggested that the original meta-analysis results were likely to be an overestimate of the true OS and DFS effect-sizes but strengthened the belief that MYC-N is a potentially important prognostic marker in neuroblastoma.

CONCLUSION

Sensitivity analyses in meta-analysis allow more appropriate and reliable conclusions when problems such as unavailable estimates and dissemination bias are present.