The first thing, and this is not just related to clinical trials, but hopefully to this disease as whole, is one of nomenclature. We have wanted to move beyond the term castration because it's not a particularly friendly term for either patients or caregivers alike. It was developed in the context of Prostate Cancer Working Group 2 when we only had ADT, and so everything oriented around serum testosterone levels. But now we have androgen synthesis inhibitors. We have really potent anti-androgen receptor drugs. We have new drugs that manipulate the AR in a variety of ways so we can move from a strictly endocrinologic definition to now a mechanistic one. And so we will be using androgen pathway modulator as the core root terminology that's naive, which is APMN, sensitive APMS, and resistant APMR.
And that transition from APMS to APMR is really progression with ADT alone, or ADT and ARPI alone, or as a doublet, or as a triplet with another drug, or some future novel AR targeting agent. Furthermore, APMR will now be designated according to prior therapy as opposed to first line, second line, third line. So you need to say PARP exposed, lutetium exposed, taxane-based chemotherapy exposed, because in today's day and age, first line, second line, third line doesn't really tell us what's going on with the patient.
Furthermore, in today's day and age, you need to specify what the genotype is or the HRD. If so, BRCA-positive, -negative, what's the mutation? MSI-high or not? The phenotype needs to express, do they have any PC? Do they have double-negative disease? Do they have standard adenocarcinoma? And as just mentioned, the imaging modality used to define their extensive disease needs to be specified whether that's PET, CT, MRI, bone scintigraphy, et cetera.
So some mention has been made of Working Group 3 criteria, and we did have a criteria that served us really quite well for a very long time. We controlled for flare with a two plus two rule, which indicated that at the start of treatment, you wouldn't declare the patient a progressor until they had at least two new lesions on the first on-treatment scan and then another, at least two new lesions on the subsequent scan. And then after the flare period, we'd had two plus zero rule, which implied that if you had two new lesions, at least then you had to wait for a confirmatory scan and then backdate the data progression to the original scan with the two new lesions. And as you can see, with that confirmatory scan, you can have a significant aggregation in the volume of disease.
On the other hand, this definition, despite its flaws in terms of that confirmatory scan and despite other issues that some might have, had a really reliable, robust cross-clinical trial, cross mechanism of action association between that definition of rPFS and overall survival, all of which were established through randomized prospective trials with clinical endpoints. And that association is more or less around 0.7.
And as Will showed you, this definition has served us really, really well in terms of drug development. This is a redo of the slide that Will had shown showing the last 10 years of drug approvals, and I put a check mark around those which had an rPFS endpoint that related to the approval of that drug. I would say that for all of the sponsors here that might object to incorporating PSMA into clinical trials as a result of cost concerns, think of what the cost of not doing it would be, because without interim endpoint, all of these trials would have to wait for OS that you see that are now drugs that are approved on the basis of rPFS. On the other hand, you don't want to really change the timing of rPFS to the first indication of a PSA rise on a clinical trial because you will move that rPFS definition to very, very early and probably then really erode the association between rPFS and OS.
The goal is to improve patient safety without changing the timing of the rPFS endpoint, at least not without data. We have then for Prostate Cancer Working Group 4 eliminated the confirmatory scan for those patients who have a high volume of rapidly progressive disease, which we set a cutoff of six or more lesions. So instead of waiting for that confirmatory scan and then backdating it to that six or more progressive event, instead we will then just declare progression on the date of the six new lesions having been identified that should not change the date of rPFS, whereas in the lesser disease or oligoprogressive context of five or fewer lesions where a change in systemic therapy might not be in order where anti-metastasis-directed therapy might be in order or where generally you don't really have to declare progression quite so early, the usual rules will apply.
Now, in terms of approaching PSMA PET and incorporating that into Prostate Cancer Working Group 4, we basically set out basic set of principles. Endpoints need to be simple, feasible, non-proprietary in order to avoid planned obsolescence as software generally is. Require clinical validation, so just like Working Group 2 was a call for evidence and validation, this two is as well. And that a reaffirmation that progressive is a regulatory and radiographic endpoint and does not imply it's the end of treatment benefit, and does not imply that the patient must change treatment at that time that the progressive event is declared.
We know we have several other schema in terms of RECIP, PPP, other people just eyeballing the disease and deciding it's bigger and brighter. These are widely acknowledged as non-validated and not ready as a standalone, there's no cutoff that could serve as a standalone clinical trials endpoint, and therefore we're left really with two things, complete clearance below blood pool of the PSMA PET, which we're going to call a CR, and then progression. And because we don't know what the thresholds based on volume or SUV are in terms of declaring a progressive lesion, we're going to consider progression as a new lesion. That's going to be based on the intensity, based on the pattern of spread, and based on the anatomic correlates. And with those three things, the reader needs to have a high confidence of declaring it a new lesion. Otherwise, what's the rush? If you can't really say that it's a new lesion, keep the patient on, and let's be clear that it should be.
A number of sponsors have proposed new ad hoc combined criteria looking at molecular imaging and anatomic imaging and bone scintigraphy and serum biomarkers. We highly discourage such efforts. Keep these as separately reported because it's important to be able to compare and contrast what these endpoints are.
Now, for PSMA PET, we have separated progression in bone nodes and pulmonary mets from those in liver and other non-pulmonary mets. So any patient who is progressing with more or six or more lesions, same rule of five applies, that progression has occurred at that date, no confirmatory scan is needed. On the other hand, if just a handful of lesions have appeared, then you can afford to keep the patient on for a confirmatory scan. Obviously, non-pulmonary visceral metastases have a entirely different approach. The moment that you see by PSMA PET a convincing liver met or other non-pulmonary visceral met, it's in the patient's interest to declare that a clinical event.
MFS, I think that Will has probably covered better than I am. Working group three deferred to the regulatory authorities for approvals for apa, enza, and daro in the non-metastatic CRPC context. We're going to need a lot more input from the agency in terms of defining what's a relevant event for MFS with PSMA PET, but I just echo Will's issue that he brought up, which was you need to be sure that a lesion by PSMA PET alone is the same as a lesion that has an anatomic correlate or a bone scintigraphic correlate as well.
So how will Working Group 4 influence design? I have 24 seconds left. We have new nomenclature. We have a safer definition of rPFS without changing the stated time of rPFS to preserve all of the work that we've done so far, but also to keep patients safe. It does include PSMA PET imaging with a focus on new lesions, the rule of five that applies to bone, lung, and nodal mets, but that does not apply to non-pulmonary visceral mets and MFS will treat cautiously for the PSMA PET-only disease. Thank you very much.