LONDON, 20 Sep (APM) - Almost half of the clinical trials used to support European approval for cancer drugs between 2014 and 2016 were at high risk of bias based on their design, according to a study.
, published on Thursday in the British Medical Journal, looked at 54 trials that were used as evidence by the European Medicines Agency to support the approval of 32 cancer drugs between 2014 and 2016.
Of these, 41 (76%) were randomised controlled trials and 13 (24%) were either non-randomised studies or single arm studies.
Overall, 19 randomised controlled trials (49%) were judged to be at high risk of bias for their primary outcome, claim the authors of the study, which was led by Dr Huseyin Naci, assistant professor of health policy and Harkness fellow at the London School of Economics (LSE).
This judgement was made based on looking at the design characteristics of each study and evaluating them compared with the Cochrane tool for assessing the risk of bias. This takes into account the randomisation process, deviations from intended interventions, missing outcome data, measurement of the outcome and selection of the reported result.
The study also looked at reporting adequacy, including completeness and consistency of information in trial protocols, publications, supplementary appendices, clinical trial registry records and regulatory documents.
Key findings included that just 10 randomised controlled trials (26%) measured overall survival as either a primary or coprimary endpoint, with the remaining trials evaluating surrogate measures such as progression free survival and response rates.
There were concerns about missing outcome data for 10 studies and concerns regarding the measurement of the outcome in seven studies.
The study authors said in accompanying blog post
on the BMJ that these shortcomings are concerning as they can "produce biased findings, exaggerating the magnitude of benefit associated with the treatment under investigation".
They added that these concerns are "especially pertinent in cancer" as most cancer drugs enter the market on the basis of their effects on surrogate endpoints, rather than clinical outcomes that matter to patients, such as overall survival.
They highlighted several cancer drugs that appeared effective on surrogate endpoints but turned out to have little effect on overall survival, including Roche's Avastin (bevacizumab) for glioblastoma, Pfizer's Inlyta (axitinib) for advanced renal cell carcinoma, Novartis' Afinitor (everolimus) for breast cancer and Roche's Tecentriq (atezolizumab) for urothelial cancer.
The authors also said that some cancer drugs that appear effective on surrogate endpoints may actually have detrimental effects on survival, singling out the BELLINI trial, in which patients with multiple myeloma who received AbbVie's Venclexta/Venclyxto (venetoclax) had shorter survival than those who received a control treatment.
The authors acknowledged the difficulties of carrying out trials in cancer but issued a series of recommendations to help improve the situation. These include devising and testing novel strategies for collating and communicating information about the validity of clinical studies in the future.
This could mean the European Clinical Trials Register require the submission of a risk of bias assessment alongside completed trial results. Trial investigators could also complete a risk of bias assessment to accompany the publication of their trial reports in the peer-reviewed literature, recommend the study authors.
The BMJ also featured an accompanying editorial piece
from Dr Barbara Mintzes, associate professor at the University of Sydney, and Dr Agnes Vitry, senior lecturer at the University of South Australia, that looked at study in relation to health technology assessment (HTA) and paying for drugs.
"Poor quality clinical trials and uncertainty about patient relevant outcomes create even greater challenges when a treatment’s cost effectiveness is being compared with existing care," they said, adding that Australia's Pharmaceutical Benefits Advisory Committee, which makes national funding recommendations, frequently rejects cancer medicines because of uncertain clinical evidence.
They also pointed out data from Germany's HTA body IQWiG, which assessed 82 new cancer drugs and indications from 2011 to 2017, and found 54% to have considerable, major, or minor benefits, but 39% had no proof of added benefit, 2% did worse than comparators, and for 5%, evidence was inadequate to quantify benefits
"Uncertainty and exaggeration of the evidence that supports approval of cancer drugs causes direct harm if patients risk severe or fatal adverse effects without likely benefit, or forgo more effective and safer treatments," said Mintzes and Vitry.