Steve’s Strategic Guide to Phase II Cancer Clinical Trials
The Classic Phase II Trial: Primary Objectives
Phase II trials test the ability of the treatment to produce measurable tumor shrinkage in a small to medium sized group of patients (typically 20-100), all with the same kind of advanced cancer. A Phase II trial is the first test directed at any measure of efficacy.
The specific statistic measured is the response rate. Patients who achieve at least a 50% reduction in the total size of their mesaurable tumors are considered to have responded. If some tumor remains, it is considered to be a partial response (PR), but if no detectable tumor remains it is called a complete response (CR). Both CRs and PRs can relapse, but typically CRs last longer than PRs. Regardless, on the whole some treatments create more lasting responses than others. The exact criteria for PR and CR can vary with the trial, and also the type of cancer. For instance blood cancers which don’t form individual measurable tumors have different criteria for PR and CR.
Patients are often quoted response rates for both experimental and standard treatments without being told what the response duration might be. This is a positively key question which you should always ask. Short lived responses may mean little or nothing in terms of long term patient survival, but long term responses can indicate some patients are getting a major benefit! Response durations in journal articles are typically reported in months with numbers like 2, 7, 12+, 32+. The ‘+’ means that the patient has not yet relapsed. Large numbers with a plus like “32+” represent hope! On the other hand, small numbers like “3+” merely mean the follow-up time on that patient is short.
The Classic Phase II Trial: Standard Design
One Stage Design: In the simplest Phase II trial, a pre-determined number patients with the same type of cancer are given the treatment at the dose determined in the prior Phase I trial, and the response rate is measured.
Two Stage Design: Many Phase II trials conducted in two stages. The idea is to avoid giving patients a treatment as soon as it can be known that the treatment is ineffective. In the two stage design, after a pre-determined number of patients have been treated, the trial is paused, and the response rate is evaluated. If the response rate is less than a prespecified minimum goal, it’s concluded that the treatment is not worth pursuing and the trial is ended. Otherwise, the trial is restarted and a pre-determined number of additional patients are accrued to permit determinination of the response rate to the desired accuracy. If you are considering a Phase II trial with a two stage design, it’s worth asking if the trial has reached the second stage, since if it has, it means at least some of the patients are responding, though the response rate could still be low. A participating doctor might also know how exactly how many responded in the first stage.
Many, but not all, Phase II studies are multi-center studies, so you may have choices as to where to get treated.
Phase II Trials: Variations
Randomization in Phase II Trials: While the classic Phase II design is not randomized, a noticeable minority of real world Phase II trials are randomized. In some cases, the goal of the trial is to compare response rates with two different, but usually related, treatments. For example the trial might compare the response rate with the combination of “Drug X” and “Drug Y” to the response rate of “Drug X” alone. From the patient point of view, the strategic considerations in such a trial are basically identical to the strategic considerations for a Phase III randomized trial (See Steve’s Strategic Guide to Phase III Trials for suggestions on Phase III trial strategy).
A few Phase II trials randomize aspects of the trial which insignificant to the patient. For instance, one randomized Phase II trial was only randomized with respect to whether a core biopsy of the tumor was done before or after treatment. The idea was to let researchers compare pre and post treatment biopsy specimens, but one wouldn’t expect this to have any effect whatsoever on the effectiveness of the treatment. You need to pay close attention to just exactly what aspect of the treatment is being randomized in a randomized Phase II trial to decide how it affects your evaluation of the trial!
Adjuvant Pilot Studies: Some pilot studies of new adjuvant therapies are classed as Phase II trials. Adjuvant trials are a completely different beast from Phase II trials for advanced disease (For more on adjuvant trials, see Steve’s Strategic Guide to Adjuvant Trials).
Phase I/II Trials: Phase I/II trials combine a Phase I and a Phase II trial of the same treatment into a single protocol. First the Phase I part of the trial is done, to determine the Maximum Tolerated Dose (MTD). Then, more patients are treated at the MTD in the Phase II part, which follows immediately afterwards. The Phase I and Phase II parts are basically ordinary Phase I or Phase II trials, so you can analyze the trial as either a simple Phase I or Phase II trial, depending whether they’ve reached the Phase II part of the trial or are still in the Phase I part. The only way to find out is to talk to the investigators.
Key Eligbility Rules for Phase II Trials
- The Classic Phase II Trial Requires You Have Advanced Cancer
If there is a treatment which has a significant chance of curing you, for ethical reasons you’ll be required to have tried that treatment first. Most advanced cancers do not have a satisfactory treatment, so in most cases there isn’t a requirement to have tried standard therapies first. As with Phase I trials, if the trial is actually a variant on standard treatment you’ll be required to have yet not tried the standard treatment.
- Each Phase II Trial is Limited to a Specific Type of Cancer
In order for the results say anything potentially useful about the effectiveness of the treatment, the response rate must be determined for specific cancers, so the treatment is tested in a separate Phase II trial for each cancer in which it might be useful. There can be several Phase II trials going on at the same time, each for a different cancer. If initial Phase II trials are successful (or if more funding becomes available), additional Phase II trials may be initiated in other cancers. This means there may be results from prior Phase II trials available to consider as evidence. Some individual Phase II trials do allow any of a group of biologically related cancers, rather than just one cancer.
- Patients in Phase II trials Must have Measurable Disease
Measurable disease means that you have at least one tumor of a minimum size whose borders can be clearly visualized on x-rays and whose shrinkage can therefore be accurately measured. As an example, lung nodules are typically considered measurable, and pleural effusions are not. You only need one measurable tumor to have measurable disease: If some of your tumors are measurable, and others are not, you are still considered to have measurable disease. Because this matter is technical and changes with improving technology for visualizing tumors, I strongly suggest asking your doctor whether you have measurable disease. The exact definition of measurable disease actually varies with the trial but in most cases a lesion which is clearly measurable would be accepted as such by most trials.
Phase II Trials: Evidence and Strategies
I believe that Phase II trials can be a particularly good choice if you have advanced cancer and are looking to get cutting edge treatment. Unlike Phase I trials, you will get the treatment at what is believed to be the best dose and schedule, and unlike Phase III trials, being randomized to a treatment you don’t favor is not an issue. As always though, the choice depends on the particulars.
It’s important to realize that most new treatments in clinical trials don’t actually prove to be an advance, so I advise caution when it comes to novel treatments for which there is little evidence. If there is a known treatment which offers even a small chance of a cure or meaningful long-term remission, then unless the evidence for a truly novel treatment in trial is unusually promising, I recommend you try the known treatment first. Often there won’t actually be a treatment good enough to justify trying before entering a clinical trial of a truly novel treatment, but it’s important that you find out! If the standard treatment can already produce long term benefit for some patients, and there isn’t anything which looks like a breakthrough in trials, a Phase II trial which attempts to improve the standard by adding another drug could be a reasonable option for first line treatment. If you’ve tried standard treatments to no avail, or if standard treatments are really not worthwhile, then, of course, you don’t need as much evidence to make a Phase II trial the best choice.
The key to evaluating the promise of a Phase II trial is in the prior results. The most applicable and useful data are data from using the treatment against your kind of cancer. The amount of prior data available will be highly variable, ranging from brand new treatments just out of Phase I which have only been tested on a few patients, to situations where hundreds of patients have been treated with similar or even the same treatment with good result.
Phase I Evidence for Phase II Trials
The Phase I trial for the same treatment may have been published or presented at a meeting, and the doctors involved with the Phase II trial should certainly know the results of the prior Phase I. The data from the Phase I trial will definitely give you a good idea of what the side effects are likely to be. Data on efficacy from a prior Phase I trial is usually sketchy because few patients will have been treated at the Maximum Tolerated Dose (which is usually the most effective dose, and the dose chosen for Phase II), because patients with varying types of cancer may have been treated, and because patients in Phase I trials may not be evaluable for response due to lack of measurable disease. Still, most treatments that have gone on to be approved or used as standard treatment did have responses in at least some of the Phase I patients (See for example, Daniel Von Hoff and Judith Turner: Response rates, duration of response, and dose response effects in Phase I studies of antineoplastics. Investigational New Drugs 9:115-22, 1991). I believe you should avoid Phase II trials where the only evidence is the prior Phase I trial and there was no evidence of patient benefit in that trial. If the data from Phase I is startling, suggesting a breakthrough, that is probably enough to make the trial an excellent choice.
Prior Phase II Evidence for Phase II Trials
There may be evidence from other Phase II trials. Sometimes a second Phase II trial is done to repeat promising results before moving on to Phase III. Such a confirmatory trial would be an especially good bet! Also results may be in from Phase II trials in other cancers. While these aren’t as applicable, good results in several types of cancer could indicate the treatment has relatively broad spectrum efficacy. With some highly targeted rationally designed treatments, there may be a powerful biological rationale to suppose the treatment will work on tumors with a specific biological characteristic, regardless of the specific type of cancer. If your tumor has the appropriate characteristic, and there are good results in other patients whose tumors have that characteristic, then the trial could be a good bet, even if this is the first test in your specific cancer. Anti-angiogenic therapies, which target tumor blood vessels, may also be relatively less sensitive to tumor type, although the efficacy of such a treatment might depend on the nature of the relationship between your tumor and its blood supply, which in turn could again depend on the type of cancer.
Finally, there may be results of trials of closely related treatments. If a combination is being tested, maybe there are promising results for similar combinations. If a combination Phase II trial uses just one experimental drug, maybe the results of that drug alone are promising, and the trial you are looking at is an attempt to make it even better by adding other drugs to the treatment.
Phase II evidence will be in terms of response as above. As I mentioned above, it’s very important to ask yourself if responses are holding, or whether the responding patients are relapsing. Follow-up times may be quite short, so information on reqponse duration may be quite limited, but if all the patients have relapsed despite short follow-up, it’s obviously a bad sign! If variations of the treatment have been in test for a relatively long time there may be longer follow-up.