Unpacking Cancer Trials: Phase 1 vs. Phase 2/3 Patient Mix

Hey there! Let’s dive into something pretty important in the world of cancer research: who actually gets into those crucial clinical trials. You know, the ones testing new treatments. We often hear that the folks in trials aren’t always quite like the patients doctors see every day in the clinic – they might be younger, healthier, or have fewer other health issues.

We’ve actually looked into this before, comparing patients in the big, final-stage trials (the ones that get drugs approved) with patients getting treatment in the real world here in Italy. And guess what? There were indeed differences. Real-world patients were, on average, older, more of them were elderly, and they often had a tougher time with their daily activities (what doctors call a worse “performance status”). The gender mix seemed okay overall, though it varied for specific cancers.

That got us thinking. If the patients in the *final* trials are different from real-world patients, what about the patients in the *very first* trials? You know, the Phase 1 trials. These are the absolute earliest steps where a new drug is tested in humans, mainly to figure out the right dose and see if it’s safe. They also look for any hints that the drug might actually work against cancer.

Why Phase 1 Trials Are a Big Deal

Phase 1 trials are super important because they lay the groundwork. They help decide if a drug is worth testing in bigger Phase 2 and 3 trials. With all the cool new targeted therapies popping up, sometimes a drug even gets approved based largely on strong results from these early phases! So, who’s in these Phase 1 trials really matters. Are the selection criteria *so* strict that they create a patient group that’s already quite different from those who’ll eventually enter Phase 2/3, let alone the real world?

That was the question buzzing in our heads. We hypothesized that maybe, just maybe, the patient characteristics would start diverging right from Phase 1 compared to Phase 2/3.

Our Little Investigation

So, we decided to dig into the data. We looked at information from the European Medicines Agency (EMA) – they publish reports on drugs they consider for approval. We gathered data on age, sex, and performance status (PS) from both Phase 1 and the corresponding Phase 2/3 trials for a bunch of anticancer treatments. We focused on treatments approved between 2013 and 2022.

We managed to pull data for 97 different treatment indications, covering 103 Phase 2/3 trials and 111 Phase 1 trials. That’s a lot of trials and a good chunk of patients – over 60,000 in Phase 2/3 and over 7,000 in Phase 1. Lung and breast cancers were the most common types in our dataset.

We crunched the numbers, comparing the average median age, the percentage of female patients, and the percentage of patients with a PS greater than 1 (meaning they needed some help with daily activities) between the two trial phases for the same indications. We used a fancy statistical method to account for the different numbers of patients in the trials.

What We Found: The Age Story

Turns out, when we looked at the overall picture across all the different cancers and drugs, the difference in age wasn’t statistically significant. The average median age in Phase 2/3 trials was about 60.7 years, and in Phase 1, it was 59.7 years. That’s only a one-year difference, and statistically, it was right on the edge of being significant (p=0.051). So, *overall*, patients weren’t dramatically older or younger depending on whether they were in Phase 1 or Phase 2/3.

However, when we zoomed in on specific cancer types, we saw bigger differences, sometimes in one direction, sometimes in the other. For skin and breast cancers, patients in Phase 2/3 trials were noticeably older (about 4.2 and 3.1 years older, respectively) than those in Phase 1. But for colorectal and upper-gastrointestinal cancers, patients in Phase 1 were slightly older. No big differences emerged when we looked at different types of drugs.

The Sex Balance

Next up, we checked the percentage of female patients. Again, overall, there was no statistically significant difference between Phase 1 and Phase 2/3 trials. The average rate of female patients was 40.8% in Phase 2/3 and 45.7% in Phase 1. This is good news, as it suggests that *within the clinical trial pipeline*, women aren’t being significantly underrepresented as trials progress from early to later phases. This is important because men and women can respond differently to treatments. Having good representation early on helps us understand these potential differences better.

Just like with age, we saw some differences when looking at specific cancers. For skin cancer, Phase 1 trials had a higher percentage of female patients compared to Phase 2/3. For upper-gastrointestinal cancers and trials using cytotoxic (traditional chemo) drugs, Phase 2/3 trials had a higher percentage of female patients.

Performance Status: How Patients Feel

Finally, we looked at performance status – basically, how well patients can carry out daily activities. We compared the rate of patients with a PS greater than 1 (meaning they need some assistance). This is often a key factor in trial eligibility.

The results here were also not statistically significant overall. The rate of patients with PS > 1 was low in both groups: 2.3% in Phase 2/3 and 1.8% in Phase 1. This isn’t surprising, as clinical trials, especially early ones, often require patients to be quite well (PS 0 or 1) to ensure they can handle potential side effects and trial procedures.

Again, looking at subgroups showed some variation, but nothing statistically significant. For colorectal cancer trials, Phase 1 actually had a higher rate of patients with PS > 1. For trials with cytotoxic agents, Phase 2/3 had a higher rate.

So, What Does It All Mean?

Here’s the scoop: Our analysis suggests that, for the characteristics we looked at (age, sex, PS), the patient populations in Phase 1 trials aren’t statistically different *overall* from those in the corresponding Phase 2/3 trials that led to drug approval.

This might sound a bit counter-intuitive at first, especially since Phase 1 trials are known for having really strict eligibility criteria. You’d think they’d be *even more* selective than later phases. And they are, in terms of things like specific molecular markers, organ function, and which other medications you can take. But when it comes to these broader characteristics like age, sex, and general performance status, the selection seems to be pretty consistent from the get-go.

Why? Well, it could be that the criteria established for Phase 1 trials – which are designed to be super cautious because it’s the first time a drug is widely tested in humans – pretty much carry over into the later phases. If a drug is initially tested and seems safe and effective in a specific group (say, patients with good PS), researchers might stick to similar criteria for Phase 2/3 trials out of caution or simply because that’s where they have data. This creates a sort of “homogenization” of the patient population throughout the trial pipeline.

The Crucial Takeaway

While our findings are encouraging in the sense that patients aren’t being *further* selected *between* Phase 1 and Phase 2/3 based on these characteristics, it highlights a different point. If the selection criteria are already quite conservative and stringent in Phase 1 (which they are, for good reasons like safety), and these criteria largely persist into Phase 2/3, then the gap between *any* clinical trial population and the *real-world* patient population starts right at the beginning.

This is crucial because Phase 1 trials are becoming increasingly important, sometimes even leading to accelerated approvals. If these early trials aren’t representative of the broader patient population who will eventually use the drug, we might miss important information about how the drug works (or doesn’t work) or its safety profile in older, less well, or otherwise different patients.

A Note on Limitations

Of course, our study wasn’t perfect. We could only compare data for a limited number of indications because we relied on publicly available reports linked to Italian drug registries. Also, we had to exclude some trials that included multiple cancer types without breaking down patient characteristics for each one, which could potentially skew results due to the natural differences in who gets certain cancers.

Wrapping It Up

So, the big picture? It seems the patient selection strategy, at least concerning age, sex, and performance status, is pretty consistent across Phase 1 and Phase 2/3 cancer trials. This suggests the potential for limited generalizability to the real world isn’t just a Phase 2/3 issue; it’s baked into the process from Phase 1. This makes designing Phase 1 trials with broader patient inclusion even more vital for the future, ensuring that the promising results we see early on are more likely to translate into benefits for the diverse group of patients who need them in the real world. It’s all about making sure our research truly reflects the people we’re trying to help!

Source: Springer