A Close Look at Clinical Trials in Oncology and Beyond: State of Affairs and Industry Progress

In this episode of OncChats: A Close Look at Clinical Trials in Oncology and Beyond, Richard D. Carvajal, MD, of Northwell Health and Northwell’s Feinstein Institutes for Medical Research, discusses the progress made in cancer clinical trials, with nearly 20,000 trials worldwide and over 80 FDA-approved cancer drugs since 2015, while also noting the challenges posed by the increasing complexity of therapies and trial protocols.

Brennan: Speaking of cancer, I know you attended the ASCO Annual Meeting, which is a very largely attended conference in the oncology space, what do you think is the current state of cancer trials? I know that’s a loaded question.

Carvajal: Dr Brennan, I am so excited about what we're doing in the cancer space. I can’t even tell you. If you look globally, there are nearly 20,000 cancer trials ongoing. When you look at what the FDA has done, since 2015, almost one-quarter of the drugs that have been approved have been cancer drugs; so that’s more than 80 drugs. All of this progress has led to improvements in outcome; we’ve decreased the death rate of cancer by one-third since 1991. In part, it’s because of what we’re doing [with these trials]. I think [when it comes to] the state of clinical research in cancer, we’re making a lot of progress. As in every other field, though, it is becoming harder and harder to do these trials and deliver these sorts of therapies to patients, in part because the therapies are getting more complex.

We’ve gone from chemotherapy trials to targeted therapy/precision medicine trials, to checkpoint blockade. Now, we’re launching all these cell therapy trials with all the complexity associated there; [there are also] theranostics trials. All of this requires a tremendous amount of institutional resource and time to make it all happen. If you look at our protocols, and I’m sure this is the case for all of them, but now, they’re routinely 300 pages. I guess I’m a little bit old now, but when I started doing this in [the early 2000s], a reasonable protocol could be 50 pages. I think that just shows the complexity.