Bridging Clinical Trials and Real-World Evidence in ALK-Positive Non–Small Cell Lung Cancer Treatment
Key Takeaways:
- Real-world data show shorter time-to-treatment discontinuation (TTD) for ALK inhibitors compared with progression-free survival (PFS) reported in clinical trials, highlighting differences between controlled trial settings and routine clinical practice.
- Only about 25% of patients transitioned to second-line ALK inhibitors, suggesting that toxicity, comorbidities, and real-world patient complexity may limit treatment sequencing.
- Findings suggest real-world evidence can complement clinical trial data, helping clinicians and pathway developers make more informed decisions about therapy selection, tolerability, and treatment value in diverse patient populations.
A study examining treatment patterns in advanced ALK-positive non–small cell lung cancer (NSCLC) provides insight into how targeted therapies perform outside of clinical trials. In this interview, health economist Rahul Mudumba, PhD, MHS, of the University of Southern California discusses findings from his study “Real-world treatment patterns and time-to-treatment discontinuation among advanced ALK-positive non–small cell lung cancer patients.” The research explores real-world persistence with first-line ALK inhibitors, factors influencing treatment sequencing, and how real-world evidence can complement clinical trial data to inform value-based treatment decisions in oncology.
Please introduce yourself by stating your name, title, organization, and relevant professional experience.
Rahul Mudumba: Hi, everyone. I'm Rahul Mudumba. I'm a health economist and a PhD candidate at the University of Southern California. My research focus and interests span oncology and investigating what is effective in the real world, not what is efficacious in clinical trials or ideal settings. We herald the clinical trial as gold standard evidence, but it doesn't tell the whole story and it doesn't reflect heterogeneous patient populations.
I’m interested in how we can measure the true effectiveness in the real world and how we can better quantify that in terms of value to society and to patients. My work looks at real-world evidence and simulating cost-effectiveness over long periods of time and improving methodology for which we could do both. That led me to the project we're going to be discussing today, which explores targeted therapies for ALK-positive NSCLC.
Can you give some background about your study and what prompted you to undertake it?
Rahul Mudumba: I was investigating the decision problem of first-line therapies in NSCLC lung cancer or ALK positive NSCLC. Typically, to determine the most valuable therapy, we evaluate cost per quality through decision analytic models, such as cost effectiveness models. I was building one to evaluate which therapy was the most valuable in a first-line setting, and I came across gaps in the literature. It was not so obvious what costs were to patients and to the health care system in the first line or second line. What happens to patients when they progress? Notably, did they receive a targeted therapy in second line? Did they receive chemotherapy or some combination therapy, and what happens?
Clinical trials didn't tackle this question head on as much, and a few of the observational research studies were a bit outdated. Also, newer treatments have entered the market since then, notably brigatinib, lorlatinib. Since we only had data of alectinib and crizotinib the first-line setting, it led to a lot of questions as to what happens in the real world if a patient progresses in the first line.
Given the gaps in the literature, I tried to use some of these inputs in my study and looked at this rich data set, which is Optum claims data. With this information, I figured out the solution with regard to monthly costs, efficacy, how long the patients stay on therapy, how long they survive, and what those treatment patterns look like.
Real-world median TTD was shorter than progression-free survival reported in clinical trials, especially for crizotinib and brigatinib. How should clinicians interpret this gap when selecting first-line ALK inhibitors?
Rahul Mudumba: One caveat is that TTD is not the go-to clinical measure. Clinical trials look at progression-free survival and overall survival. You could glean overall survival from claims data, but it's hard to tell when a patient has progressed by looking only at administrative claims. A proxy for that could be a TTD when they stop taking the drug.
In the real world we're trying to find progress and the easiest proxy you compute this in a quantifiable way tends to be TTD. It's a quantifiable, easy traceable way to proxy progression-free survival. It is also shorter than progression-free survival in the clinical trial. The question was to what degree would we see it?
In the real world, you have a sicker patient population that faces more comorbidities than in a clinical trial. For clinical trials, researchers are trying to get a very clean comparison, rightfully so. We want to see how efficacious the drug can be in an ideal setting, but the tria; doesn't represent the broad patient population in the real world in which patients have a lot of comorbidities and come from different races and backgrounds.
Once you have other comorbidities or mediating factors in the real world, you can't expect the same efficacy to hold that it would in the trial. That is why there is a big gap between real-world effectiveness and clinical trial data. The broader population is probably sicker on average.
Only about 25% of patients transitioned to a second-line ALK TKI. What factors may be limiting treatment sequencing in real-world practice?
Rahul Mudumba: Building off of the previous of patients being sicker in the real work, a lot of factors come into play. One is the adverse event likeliness or the toxicity that these drugs have. If patients have other comorbid conditions and can't deal with the toxicity that these drugs produce—notably lorlatinib, regatinib, electinib—there are tradeoffs here.
For instance, oftentimes with more efficacy comes more toxicity, especially in the case of lorlatinib where you have neurocognitive toxicities that are not as evident in a trial setting.
Real world heterogeneity plays into the persistence and discontinuation rates. Persistence is lower and discontinuation is higher oftentimes because of toxicity.
