A new paper in the Journal of the National Cancer Institute, published by Oxford University Press, finds that almost 20% of patients in middle-stage cancer drug trials receive treatment that eventually prove effective enough to get FDA approval. This may have important implications for drug development and clinical trial recruitment.
The development of new medications typically has three stages. In phase 1 trials, researchers assess drugs for safety and dosing ("What is the best tolerated dose for the patient?"). Phase 2 clinical trials determine whether a new drug shows signs of efficacy ("How much does the drug reduce disease?"). Drugs showing promise in phase 2 are then evaluated in phase 3 trials, which are conducted at multiple centers with at least several hundred patients and often inform whether the FDA approves of the drug for use as treatment.
However, unlike initial phase 1 trials, all patients in phase 2 trials receive drugs at the doses intended for routine use with the expectation the drug will be active and beneficial, though this is unknown from prior experience. With more patients treated at an active dose, a drug is more likely to have side effects. Therefore, the question of efficacy in phase 2 trials is an important consideration for patients wishing to receive newer treatments.
The aim of the study was to estimate the proportion of cancer patients in phase 2 trials who receive new interventions that are later deemed safe and effective and approved by the FDA.
The researchers here identified 2,730 phase 2 clinical trials, beginning between Nov. 2012 and Nov. 2015, 1,154 of which met eligibility and 400 of which were randomly sampled for inclusion. These phase 2 trials had a total patient enrollment of 25,002 patient-participants in 608 specific cohorts or arms (including control arms) which tested 332 drugs. Twenty-five drugs were evaluated in more than 10 trials, 155 in 2-10 trials, and 152 in one trial only.
The investigation found that the FDA later approved 71 drug regimens in the tested indication, and 1 in 6 (16%) patients in phase 2 trials received such regimens. However, the authors of the study point out that accessing a drug that later gets FDA approval doesn't necessarily mean the patient will benefit. "Keep in mind that approved drugs don't work for every person. For many FDA approved cancer drugs, only half to a tenth of patients meaningfully benefit," said the paper's lead author, Charlotte Ouimet.
The authors of the study say that these findings give patients important information on expectations for phase 2 trials. Most patients who enroll in phase 2 have advanced cancers that lack standard treatment options. Many such patients are likely to regard a 16% probability of receiving a drug that later secures FDA approval for their condition as favorable odds.
"If you're a cancer patient and you are invited into a phase 2 trial, keep in mind that five of six patients get treatments that do not go on to approval" said Jonathan Kimmelman, another author. "Just the same, your odds are dramatically better than they would be for a phase 1 trial."
"Your odds of receiving an effective drug in phase 2 are quite a bit higher than for phase 1 trials. But probably still a lot lower than for phase 3 trials, where the odds are probably more like 1 in 3," Kimmelman continued.
The paper, "Proportion of Patients in Phase 2 Oncology Trials Receiving Treatments that are Ultimately Approved," is available (at midnight on February 25th) at https://doi.org/10.1093/jnci/djaf013 .
Direct correspondence to:
Jonathan Kimmelman
Professor of Equity, Ethics and Policy
McGill University School of Population and Global Health
2001 McGill College Avenue
Montreal, QC H3A 1G1, CANADA
