Industry Funding of Oncology Randomised Controlled Trials: Implications for Design, Results and Interpretation.

AIMS: Most randomised controlled trials (RCTs) in oncology are now funded by the pharmaceutical industry. We explore the extent to which RCT design, results and interpretation differ between industry-funded and non-industry-funded RCTs. MATERIALS AND METHODS: In this cross-sectional analysis, a structured literature search was used to identify all oncology RCTs published globally during 2014-2017. Industry funding was identified based on explicit statements in the publication. Descriptive statistics were used to compare elements of trial methodology and output between industry- and non-industry-funded RCTs. RESULTS: The study sample included 694 RCTs; 71% were funded by industry. Industry-funded trials were more likely to test systemic therapy (97% versus 62%; P < 0.001), palliative-intent therapy (71% versus 41%; P < 0.001) and study breast cancer (20% versus 12%; P < 0.001). Industry-funded trials were larger (median sample size 474 versus 375; P < 0.001) and more likely to meet their primary end point (49% versus 41%; P < 0.001). Among positive trials, there were no differences in the magnitude of benefit between industry- and non-industry-funded RCTs. Trials funded by industry were published in journals that had a significantly higher median impact factor (21, interquartile range 7, 28) than non-industry-funded trials (impact factor 12, interquartile range 5, 24; P = 0.005); this persisted when adjusted for whether a trial was positive or negative. CONCLUSIONS: The vast majority of oncology RCTs are now funded by industry. Industry-funded trials are larger, more likely to be positive, predominantly test systemic therapies in the palliative setting and are published in higher impact journals than trials without industry support.

Five years of EMA-approved systemic cancer therapies for solid tumours-a comparison of two thresholds for meaningful clinical benefit.

OBJECTIVE: Several societies have proposed frameworks to evaluate the benefit of oncology drugs; one prominent tool is the European Society for Medical Oncology Magnitude of Clinical Benefit Scale (ESMO-MCBS). Our objectives were to investigate the extent of European Medicines Agency (EMA)-approved cancer drugs that meet the threshold for 'meaningful clinical benefit' (MCB), defined by the framework, and determine the change in the distribution of grades when an adapted version that addresses the scale's limitations is applied. METHODS: We identified cancer drugs approved by the EMA (2011-2016). We previously proposed adaptations to the ESMO-MCBS addressing its main limitations, including the use of the lower limit of the 95% confidence interval in assessing the hazard ratio. To assess the MCB, both the original and adapted ESMO-MCBS were applied to the respective approval studies. RESULTS: In total, we identified 70 approval studies for 38 solid cancer drugs. 21% of therapies met the MCB threshold by the original ESMO-MCBS criteria. In contrast, only 11% of therapies met the threshold for MCB when the adapted ESMO-MCBS was applied. Thus 89% and 79% of therapies did not meet the MCB threshold in the adapted and original ESMO-MCBS, respectively. CONCLUSIONS: In most of the cancer drugs, the MCB threshold is not met at the time of approval when measured using both ESMO-MCBS scales. Since approval status does not translate into a MCB, stakeholders and decision makers should focus on the benefit/risk ratio of anticancer drugs to assure an appropriate allocation of resources in health care systems.

Population-based study to re-evaluate optimal lymph node yield in colonic cancer.

BACKGROUND: It is well established that lymph node (LN) yield in colonic cancer resection has prognostic significance, although optimal numbers are not clear. Here, LN thresholds associated with both LN positivity and survival were evaluated in a single population-based data set. METHODS: Treatment records were linked to the Ontario Cancer Registry to identify a 25 per cent random sample of all patients with stage II/III colonic cancer between 2002 and 2008. Multivariable regression and Cox models evaluated factors associated with LN positivity and cancer-specific survival (CSS) respectively. Optimal thresholds were obtained using sequential regression analysis. RESULTS: On adjusted analysis of 5508 eligible patients, younger age (P < 0·001), left-sided tumours (P = 0·003), higher T category (P < 0·001) and greater LN yield (relative risk 0·89, 95 per cent c.i. 0·81 to 0·97; P = 0·007) were associated with a greater likelihood of LN positivity. Regression analyses with multiple thresholds suggested no substantial increase in LN positivity beyond 12-14 LNs. Cox analysis of stage II disease showed that lower LN yield was associated with a significant increase in the risk of death from cancer (CSS hazard ratio range 1·55-1·74; P < 0·001) compared with a greater LN yield, with no significant survival benefit beyond a yield of 20 LNs. Similarly, for stage III disease, a lower LN yield was associated with an increase in the risk of death from cancer (CSS hazard ratio range 1·49-2·20; P < 0·001) versus a large LN yield. In stage III disease, there was no observed LN threshold for survival benefit in the data set. CONCLUSION: There is incongruity in the optimal LN evaluation for colonic cancer. Although the historically stated threshold of 12 LNs may ensure accurate staging in colonic cancer, thresholds for optimal survival are associated with far greater yields.

Lymph node evaluation for colon cancer in routine clinical practice: a population-based study.

BACKGROUND: Guidelines recommend that 12 or more lymph nodes (lns) be evaluated during surgical resection of colon cancer. Here, we report ln yield and its association with survival in routine practice. METHODS: Electronic records of treatment were linked to the population-based Ontario Cancer Registry to identify all patients with colon cancer treated during 2002-2008. The study population (n = 5508) included a 25% random sample of patients with stage ii or iii disease. Modified Poisson regression was used to identify factors associated with ln yield; Cox models were used to explore the association between ln yield and overall (os) and cancer-specific survival (css). RESULTS: During 2002-2008, median ln yield increased to 17 from 11 nodes (p < 0.001), and the proportion of patients with 12 or more nodes evaluated increased to 86% from 45% (p < 0.001). Lymph node positivity did not change over time (to 53% from 54%, p = 0.357). Greater ln yield was associated with younger age (p < 0.001), less comorbidity (p = 0.004), higher socioeconomic status (p = 0.001), right-sided tumours (p < 0.001), and higher hospital volume (p < 0.001). In adjusted analyses, a ln yield of less than 12 nodes was associated with inferior os and css for stages ii and iii disease [stage ii os hazard ratio (hr): 1.36; 95% confidence interval (ci): 1.19 to 1.56; stage ii css hr: 1.52; 95% ci: 1.26 to 1.83; and stage iii os hr: 1.45; 95% ci: 1.30 to 1.61; stage iii css hr: 1.54; 95% ci: 1.36 to 1.75]. CONCLUSIONS: Despite a temporal increase in ln yield, the proportion of cases with ln positivity has not changed. Lymph node yield is associated with survival in patients with stages ii and iii colon cancer. The association between ln yield and survival is unlikely to be a result of stage migration.

Do Contemporary Randomized Controlled Trials Meet ESMO Thresholds for Meaningful Clinical Benefit?

Background: The European Society for Medical Oncology (ESMO) recently released a magnitude of clinical benefit scale (ESMO-MCBS) for systemic therapies for solid cancers. Here, we evaluate contemporary randomized controlled trials (RCTs) against the proposed ESMO thresholds for meaningful clinical benefit. Methods: RCTs evaluating systemic therapy for breast cancer, nonsmall cell lung cancer (NSCLC), colorectal cancer (CRC), and pancreatic cancer published 2011-2015 were reviewed. Data were abstracted regarding trial characteristics and outcomes, and these were applied to the ESMO-MCBS. We also determined whether RCTs were designed to detect an effect that would meet clinical benefit as defined by the ESMO-MCBS. Results: About 277 eligible RCTs were included (40% breast, 31% NSCLC, 22% CRC, 6% pancreas). Median sample size was 532 and 83% were funded by industry. Among all 277 RCTs, the experimental therapy was statistically superior to the control arm in 138 (50%) trials: results of only 31% (43/138) of these trials met the ESMO-MCBS clinical benefit threshold. RCTs with curative intent were more likely to meet clinically meaningful thresholds than those with palliative intent [61% (19/31) versus 22% (24/107), P < 0.001]. Among the 226 RCTs for which the ESMO-MCBS could be applied, 31% (70/226) were designed to detect an effect size that could meet ESMO-MCBS thresholds. Conclusion: Less than one-third of contemporary RCTs with statistically significant results meet ESMO thresholds for meaningful clinical benefit, and this represents only 15% of all published trials. Investigators, funding agencies, regulatory agencies, and industry should adopt more stringent thresholds for meaningful benefit in the design of future RCTs.