Implementing Prostate Cancer Working Group 3 Criteria in Clinical Trials- Michael Morris and Lawrence Schwartz

February 25, 2020

Alicia Morgans welcomes Michael Morris and Lawrence Schwartz to discuss the Prostate Cancer Working Group criteria as a refresher for investigators who are treating men with prostate cancer in clinical trials. 


Michael J. Morris, MD., is the Clinical Director, Genitourinary Medical Oncology Service & Prostate Cancer Section Head, Division of Solid Tumor Oncology, Memorial Sloan Kettering Cancer Center, New York, NY.    Dr. Morris is one of the leaders of the Prostate Cancer Clinical Trials Consortium (PCCTC). This initiative is designed to increase patient access to clinical trials across the country. He is also the chair of the Genitourinary Committee of the National Cancer Institute’s cooperative group Alliance for Clinical Trials in Oncology, which facilitates testing new drugs nationwide.

Lawrence H. Schwartz, MD, is the James Picker Professor and Chairman for the Department of Radiology at Columbia University Medical Center. He also is the Chief of the Radiology Service and Attending Physician at New York-Presbyterian Hospital.

Alicia Morgans, MD, MPH Associate Professor of Medicine in the Division of Hematology/Oncology at the Northwestern University Feinberg School of Medicine in Chicago, Illinois.

Read the Full Video Transcript

Alicia Morgans: Hi, I'm excited to speak with you today about the Prostate Cancer Working Group criteria as a refresher for investigators who are treating men with prostate cancer on clinical trials. Today I'll be speaking with Dr. Michael Morris and Dr. Larry Schwartz about these criteria. Thank you for your time.

So I know investigators who are treating men with prostate cancer have many questions, particularly as they are trying to understand imaging over the course of the trial. And the two of you are really here to give us some recommendations of how to think through things in a really, sort of, clear cut and easy way. So I think we want to just start with PSA as our first point. Can you tell me a little bit about how you think about monitoring PSA when you are caring for men on prostate cancer clinical trials per the Prostate Cancer Working Group criteria?

Michael Morris: Sure. Well historically speaking, PSA used to be the primary indicator of disease progression and what would warrant a treatment change. That changed with Prostate Cancer Working Group 2 because PSA historically also didn't really serve us too well in predicting clinical benefit, like overall survival. So instead of using PSA as an indicator of progression or indicating the need for a treatment change, really the focus has now shifted to scans. So PSA is monitored but is generally not a decision-maker for an indication of change treatment nor for indicating that the patient has progressed. It's followed as a biomarker, but it is not a key decision-maker. And indeed, in many trials, you can't change treatment without incurring a protocol violation on the basis of PSA exclusively.

Alicia Morgans: So if the PSA is rising and scans remain unchanged, in trials that are following men with prostate cancer that are really adhering to the Prostate Cancer Working Group criteria, you should not change treatment based on a rising PSA alone?

Michael Morris: Correct.

Alicia Morgans: Okay.

Michael Morris: Yes.

Alicia Morgans: Very important. Certainly, clinical things could change your mind, but PSA alone is not the thing.

Michael Morris: Right.

Alicia Morgans: So when we're thinking about monitoring men with imaging, which is really the basis on which we're going to change most therapies in these studies, can you tell us a little bit about how you think about bone scans?

Michael Morris: Sure. Well, there's a real focus on bone scans because most patients will have a bony disease is the primary manifestation of their metastatic disease. It will represent most of their disease burden. And Prostate Cancer Working Group 2 and 3 have set out a relatively straightforward and simple set of criteria that have been prospectively validated now to document disease progression. We call those the two plus two criteria because the criteria are really predicated on counting to two. It doesn't require that you count all lesions, just new lesions.

So you take the patient's baseline scan, you document the presence or absence of metastatic disease, and for every scan that follows that baseline scan, that's the pretreatment scan, you're basically looking at what's new on the subsequent scan relative to the baseline scan. What's new for Prostate Cancer Working Group 3 is for what we consider to be the baseline scan. So when a patient first starts treatment, the baseline scan is the pre-treatment scan. However, once the patient gets his first on-treatment scan, the baseline scan is that scan because it's that scan that we think will have the greatest disease burden because it would be that scan in which there were any pseudoprogression or flare. That's a paradoxical increase in bone lesions or increased conspicuity of existing bone lesions will occur. So we consider that to be the baseline scan for the future scans after that first scan is performed.

So what's progression on a bone scan? Well, you can meet progression really by two different means. For patients who are just starting their therapy, if they have at least two new lesions on their first post-treatment scan or on-treatment scan, and then another minimum of two lesions on the subsequent scan, so a total of four new lesions relative to the pretreatment scan, that's progression. It's two plus two because it's a minimum of two new lesions per scan for those first two scans. We recognize that many patients will flare and so they'll get perhaps two new lesions at least on their first post-treatment scan, but that should stabilize over the course of therapy.

The second way that patients progress on a bone scan is that once the flare period is over, so once that first post-treatment scan has been completed, that scan is now the new baseline. And so if you have two new lesions on any subsequent scan after that flare period relative to the week nine scan, that are then confirmed on a subsequent scan that that is progression. Now, that can manifest in many ways. You could have two new lesions, or at least two new lesions on your week nine, that's your first post-treatment scan, and then no new lesions for every scan following. But then let's say at week 37 the patient has two new lesions relative to that week nine scan, and that's confirmed on the next scan. That's progression. You could have two new lesions on your week nine scan, then no new lesions for three or four scans, then one new lesion, then no new lesions, then one new lesion. That's progression when it's confirmed on the next scan.

That's in contradistinction to the two plus two rule, which is two new lesions on the first scan, that's week nine, and two additional new lesions on the next scan. That's week 17. That's progression as well. So it's really just counting to two. There are really just two scenarios. And it doesn't require any special equipment, doesn't even have to be a radiologist who does this because we figured that any trained person who's involved in the trial can count to two.

Alicia Morgans: So let's just talk through that just a couple more times just to emphasize a few things. So it does not matter how many new lesions the patient has on the week nine scan, correct?

Michael Morris: Exactly.

Alicia Morgans: This is considered flare. And then this week nine scan becomes the baseline to which every subsequent scan is then compared?

Michael Morris: Yes. I think that's a really important point, no one should come off for radiographic progression at week nine, period. No matter how bad that looks. That's the flare scan. That's the new baseline.

Alicia Morgans: Absolutely.

Lawrence Schwartz: The concept of the week nine-scan being the new baseline is really just to emphasize that point, that nobody should come off. Obviously nobody comes off on a baseline scan anyway, and that point just emphasizes it.

To back up a little bit, we've been doing bone scans for decades really, but as Mike pointed out, they were really assessed very much qualitatively rather than quantitatively because they were assessed together with PSA. This set of rules uses the exact same imaging modality that we've used, Michael and his teams, we've actually validated this, that it does correlate with outcomes, so now we have a ruleset which is relatively easy. I think it's actually very easy to use, especially after you've done it a few times. And this ruleset then could be used literally globally in every single clinical trial to give an identical reproducible set of results and provide us really with a lot of information moving forward about prostate cancer progression.

Alicia Morgans: Absolutely. So let's just reiterate this two plus two in the scenarios that you've described one more time with some visuals so that we can make some sense of it between what we hear and also what we see.

Michael Morris: So in scenario one, we have a patient that has early progression. We see that at week nine he has two new lesions relative to his baseline scan, and we've circled those new lesions, and then at week 17 he has two additional, or a minimum of, two additional lesions, and we've circled those in red so that you can see then that he meets the two plus two rule with two additional lesions at week nine and at least two additional lesions at week 17.

Alicia Morgans: And at what time would you say he's officially progressing by Prostate Cancer Working Group criteria?

Michael Morris: The date of progression for patients meeting the two plus two rule is the date of the second scan with two or more new lesions. So in this circumstance, it's week 17.

Alicia Morgans: Okay, great.

Michael Morris: In the second scenario, we have a patient who at baseline, you have his baseline scan, and at week nine he has two new lesions, but he doesn't come off study because he hasn't met the two plus two rule. At week 17, he's stable and in all the subsequent scans, he's stable as well. But at week 37, we see that he has two additional new lesions relative to his week nine scan, and these are then confirmed at week 49. At the date of progression is backdated to when he had the two new lesions relative to week nine appear. So it's not the confirmatory scan, it's the date of the second mew lesion. In this circumstance it would be week 37.

Alicia Morgans: Great, thank you.

Michael Morris: And while most patients' disease will primarily be manifest by a bony disease, many patients, especially with castration-resistant disease, especially those who have been treated with enza or abi, will have a significant component of soft tissue disease. And in today's day and age, accounting for progression in soft tissue disease is very important. Prostate cancer working code three addressed this head-on. And Larry, if you could walk us through what those changes are and how it contrasts with RECIST and how it's similar to RECIST. I think that would be very helpful.

Lawrence Schwartz: Sure. So with soft tissue disease in Prostate Cancer Working Group 3, we're trying as much as possible really to follow resist 1.1. So for instance, where most of the soft tissue disease will be in lymph nodes, we're following the measurement rule of measuring the short axis. So a lymph node that's greater than 1.5 centimeters in the short axis is unequivocally considered to be metastatic. Now we all know, especially in prostate cancer, that there are men that have lymph nodes that are less than 1.5 centimeters that may actually have metastatic disease. So in the category and the size between one and 1.5 centimeters, that's kind of a gray zone for us. We want to look at those lymph nodes, we want to follow them, but we don't necessarily have to consider them to be target lesions.

What our overarching goal is, is to really have a high specificity for malignancy. As a result of that, lymph nodes less than one centimeter, again, even in some cases, while they may represent a metastatic disease, we don't consider those. So when we actually measure this disease, we want to be fairly certain that it actually does represent metastatic disease. So we follow that rule for the lymph nodes. And again, that is consistent with RECIST 1.1.

We also want to look for lesions in other organs. Rarely we may see lesions in the mesentery, in the omentum, in the liver, even in the lung at times. And again, for those, we also want to follow the rules of RECIST 1.1 and any new site of disease actually does represent progression.

What we're asking the radiologist and the oncologist to do is to capture that burden of disease as much as possible. We state in the criteria, measuring or capturing, annotating up to five lesions per organ. We don't want this to be burdensome at all in terms of your clinical trial practice, but what we really do want to do is capture the patient's burden of disease. So we put this in the framework of the criteria but also in the framework of us studying disease progression so that we can understand, as we collect more and more of this data, where and how this type of disease progression, and if there's a phenotype of disease progression so that we can understand if progression in certain organs, actually are worse than in other sites of disease.

Alicia Morgans: So I've definitely heard investigators say, do you really want me to measure five lesions in the liver or five lesions in the lung? Is that really helpful? Is there any basis to that information or any use for that information other than the original lesion that's my target lesion that I'm going after?

Lawrence Schwartz: So we would really like investigators to capture, as best as possible, the patient's tumor burden. And it's very difficult to say that in a criteria so we gave a somewhat arbitrary number of five. We recognize that that in and of itself, not to overuse the word, but actually could be burdensome. Again, the goal is not to overmeasure or to cause an excessive amount of work, but really the goal is to capture the patient's tumor burden so that when we and other groups, and hopefully some of the investigators involved, go back and look at the data, we're able to distinguish between a patient who may just have one new lymph node and a patient that may have four or five lung metastases and or four or five liver metastases. We believe that there's a potential that those two patients will act very differently, and we'd like to understand that over time.

Michael Morris: We're not requiring that they measure five. We're requiring that they capture really up to that as a maximum. But we do have data that patients who have hepatic disease behave differently than lung disease, so really the effort is not to overburden the lesion counting, but instead to ensure that there's enough data capture from each organ to separately distinguish what's happening in each organ system.

Lawrence Schwartz: In fact, this may actually be the beginning of Prostate Cancer Working Group 4 and this will give us the data to really optimize that set of rules. And perhaps we wouldn't have to count it all in that setting, but just note the presence or absence of disease in some of these other organs.

Alicia Morgans: Great. So when we think about soft tissue imaging or even bone, sometimes people reflect back, well in my practice, I'm thinking about now using PSMA Axumin® scans, other scans that may be useful, are these included in current clinical trials where investigators would integrate these into their measurements?

Michael Morris: So the answer to that is no. And the reason for not including it in any molecular imaging in Working Group 3 is the lack of prospective validation for post-treatment changes on molecular images correlating with a clinical benefit. These clearly are the imaging modalities of the future, and we definitely need to do that validation just as we did with bone scans and CT with their work in Group 2 and 3. But for now, molecular imaging in clinical trials is considered to be exploratory. Most contemporary clinical trials will not use molecular imaging to make treatment changes or declare progression. And the whole effort to clinically qualify changes, post-treatment changes on molecular imaging, as reflective of a clinical benefit is an ongoing one but has so much more work to do that we would consider these to be exploratory investigational endpoints rather than definitive indices that something clinically relevant has happened.

Alicia Morgans: Okay, so not for use at this time. And so the other question that investigators ask is, do I need a radiologist to be trained in this to do all of these measurements and these assessments or is this something that I, as the investigator, as the oncologist would do on my own? How do you think about that?

Lawrence Schwartz: I think it's always optimal to have a radiologist involved in the process. We have added expertise and I think the team together, the reason we're both here together, is that we really can give the best information to the patient and inform the clinical trial the best. We recognize that that's not always possible, so certainly an oncologist who's been trained in doing this can participate in this process. Of course. I think the key is, if there are any questions at all in the process, they can reference one of their radiology colleagues in a specific clinical case within maybe a specific question.

Alicia Morgans: So encouraged but not necessarily required then?

Michael Morris: I mean, I think we recognize that many clinical trials centers are in the community, they don't necessarily have a reference radiologist for study and that there are different resources, both depending on the site that an investigator's in, and those resources can vary by country as well. For the validation studies of Working Group 2 and Working Group 3, there were formal assessments between central reads and local reads. Those assessments showed very good concordance between them and we recognize that those studies were international, representing a variety of practice types. So I agree with Larry that it's ideal to have a reference radiologist, and that's certainly the case for the soft tissue reads and the RECIST reads, but we also recognize that there are a wide variety of clinical contexts in which practitioners are participating in clinical trials and that you don't always have the benefit of that.

Alicia Morgans: Okay. So we talked a little bit about how radiographic progression is defined by Prostate Cancer Working Group criteria, what about changing treatment though? Is it always changed when we see radiographic progression?

Michael Morris: So I think one of the really important advances that Working Group 3 represents is that the criteria for documenting progression may be very different from what the clinical needs of the patient are. So there is no requirement under Working Group 3 to stop treatment simply because you have reached a protocol's criteria for progression. So a patient may, for example, have bone-dominant disease, but several small lymph nodes have increased in size, so he's met progression criteria by Working Group 3. But the patient's disease may be otherwise well controlled. The dominant burden of his disease may not be progressing. He may be experiencing good palliative relief from the treatment. There's no requirement that the patients stop therapy until it's clear that there's no more clinical benefit.

So you would simply, in that circumstance, document the progressive event for the trial's purposes but you keep the patient on study. And you could keep him on treatment until you, as a clinician, feel that his benefit has been reached in terms of receiving that treatment and it's time to change.

Alicia Morgans: Okay, so I think this concept of no longer clinically benefiting is one that is really important to emphasize that comes out of Prostate Cancer Working Group 3, you can continue the treatment on trial until you feel that that patient has both maybe radiographically progressed, but also is no longer clinically benefiting.

Michael Morris: Yeah. The purpose of this is to allow the biomarker and the endpoint of the trial to be satisfied, but to let the doctors be the doctors and decide when the treatment should end because they're the ones who are sitting in front of the patient and know how the patient is doing. Both goals can be satisfied to satisfy the trial and to satisfy the appropriate care of the patient.

Lawrence Schwartz: And ultimately, having this data, again, will help us refine criteria in the future by being able to look at the imaging, so-called post progression of our existing criteria.

Alicia Morgans: Are there any other frequently asked questions that you receive as people who have been involved in developing Prostate Cancer Working Group criteria from investigators that you think it would be interesting to explore and to sort of clear up while we have the opportunity?

Michael Morris: One question that many providers have as investigators on studies is the scanning schedule that Working Group 3 represents. So that schedule is a scanning schedule that sort of defaults to scanning in the first six months, every eight to nine weeks, and then scanning after the first six months of treatment basically every three months. There've been questions that have been raised in terms of insurers' reimbursement of the scanning in that first period of six months, basically every two months. And for some studies, those will be research non-billables. But the agency itself, the FDA has published on Prostate Cancer Working Group 2, supported the schedule, because that's the validated schedule. And what I would recommend is that that publication, which we can make available to investigators through this website, be sent to the insurer, showing that there is a regulatory endorsement of this schedule. And in order to make a compelling argument that this is the schedule in which meaningful changes have not only been validated and demonstrated in prospective studies but which also has a regulatory endorsement of.

Alicia Morgans: Absolutely. Anything from a radiologist perspective that you hear?

Lawrence Schwartz: No, but I would just echo that. And for the patient, they should feel confident that getting these scans at these intervals will actually give appropriate information to make judgments about their disease. We're not attempting to over-scan or under-scan. We always, in the future, will try to tweak the amount of scanning that we do, but that not only depends upon our criteria, that also depends upon the therapies that we're actually using. So this is going to be an evolving process. But right now this is the guideline for scanning, which has given us validated information not only for drug approval purposes but for the individual patient.

Alicia Morgans: I think particularly because we're not making treatment changes or decisions about progression on PSA, this is something that should make patients feel more confident that we will identify change if it's occurring radiographically, whether their PSA is rising or not. Because I do think that that's a fear when people see PSA rising to have that scan is also quite reassuring to the individual. And to know that it's a safety time point that's been really endorsed by the FDA is also very encouraging.

Michael Morris: Yeah, I think that's a very good point. The purpose of the criteria outlining a more aggressive scanning schedule upfront is to ensure that the rapid progressor does not remain on study too long and then slowing things down on the backend to ensure that the slow progressor isn't prematurely terminated from his treatment. And I think that these criteria satisfy that.

I think it also bears pointing out that these are not arbitrary at this point. These criteria have been validated in over 4,000 patients now, showing a very good correlation with clinical outcome. And so I think that both the investigator and the sponsor should really adhere to these guidelines because these have been shown to be safe, reproducible, and the schedule and the methodology has been shown to be an excellent indicator of overall survival.

Alicia Morgans: Wonderful. So are there any other final points that you want investigators to be aware of as we wrap up this informational session?

Michael Morris: I think it's important to recognize, especially now that we're doing so many international trials, that we do certainly appreciate that bone scans are probably a modality of the past, except in this context and the effort to clinically qualify newer imaging modalities is certainly something that's in the forefront of our minds. It will be very helpful to have a uniform set of regulatory endorsements around, especially PSMA imaging across the globe so that everyone can actually do the same imaging modalities. But it's important not to abandon bone scans for clinical trials' purposes yet because this is all we have right now in terms of a validated biomarker for overall survival.

Alicia Morgans: I think that's critically important. And although we're all very excited, and in many cases using molecular imaging techniques, those are not what we have as validated tools for prostate cancer clinical trials at this point, but that is a focus of the future. But not at this point.

Lawrence Schwartz: The great news is that we're not trying to go back to the future in terms of using bone scans. And in fact, we now have a framework with which all molecular imaging can now be validated, and we just have to move that forward. That's in the foremost of our minds to do as well with the entire global community. So in fact, we look forward to doing that.

Alicia Morgans: Wonderful. All right. Well, thank you both so much for your time and your expertise.

Michael Morris: Thank you, Alicia.

Lawrence Schwartz: Thank you.