Drug development is a lengthy and costly process with an average of 10+ years for a (successful) drug to make it from discovery to approval and a research and development (R&D) bill of over $ 2 billions. Of course, R&D programs are associated with very high attrition rates and most drugs do not make it past early phase trials where safety (primarily) and initial signs of efficacy are assessed.
For serious diseases, including cancer, there is an urgent need to accelerate the pace of the drug development and approval process and this can be achieved through improved clinical trial design and working alongside the regulators to optimise the drug application review process.
Oncology clinical trials of novel agents can be grouped into two main categories: basket and umbrella trials.
Umbrella trials
In this type of study, patients that have tumors that started in the same body part, i.e. common tissue of origin (so this can be for example breast, bladder or skin) are evaluated.
Despite sharing a common tissue of origin, it is often the case that tumors of the same tissue type harbor different molecular abnormalities. This means they are biologically driven by different mechanisms and as such are are considered to be different disease “sub-types”.
In umbrella trials, patients are enrolled entered into distinct treatment groups (arms) in the trial based on the molecular profile of their tumor; in this way a new drug targeting the particular molecular process within each disease sub-type can be tested. This type of trials is beneficial for cancer types known to be driven by many molecular abnormalities (e.g. breast cancer that is driven by HER2 (HER2+) or HER3 (HER3+) or HR (HR+)).
Basket trials
In this trial design the focus is on the molecular abnormality/mutation or biomarker that the tumor carries regardless of the tissue where the cancer originated and the trial enrols patients into a single arm of treatment targeted at the molecular abnormality in question. This design is beneficial where the molecular abnormality driving the cancer is found in many cancer types but it is not highly common within one cancer type. For example, the anti-PD1 drug pembrolizumab was found to work in cancers with high microsatellite instability (a predisposition to have a high mutation rate due to defects in DNA mismatch repair) across many indications.
In some cases, drug efficacy is determined by the presence of the molecular abnormality in a manner that is also dependent on the cancer tissue type, as was the case in the basked trial by Hyman and colleagues which tested the efficacy of the drug vemurafenib in cancers with V600E mutation in the protein BRAF.
It is important, however, to bear in mind that early clinical trials of new drugs (also known as Phase 1) are often open to all-comers, provided the patients meet certain eligibility criteria (fitness, adequate liver/kidney function etc). It is these early “signal-finding” studies that will look at the safety of the drug and allow the early assessment of signs of efficacy in the whole patient population as a whole and any subgroups of patients within it, if relevant.
This evaluation informs the next steps in the drug development process and the design of subsequent, more advanced trials (Phase 2 and Phase 3) focused on select patient groups that in early phase trials may have shown signs of clinical benefit, or the termination of the program if the drug does not meet the pre-specified safety or anti-tumor activity criteria.
With the explosion in numbers of new drugs being discovered and developed in oncology, improved trial designs that can deliver safe and effective therapies to the patients that need them is instrumental for reducing overall costs and timelines.
Reference
Le DT et al. Mismatch-repair deficiency predicts response of solid tumors to PD-1 blockade. Science 2017;357:409–13.
Hyman DM et al. Vemurafenib in multiple nonmelanoma cancers with BRAF V600 mutations. N Engl J Med 2015;373:726–36.
1 thought on “Oncology: the difference between umbrella and basket clinical trials”