I’ll start by talking about Prostate Cancer Working Group 4 which was published in The Journal of Clinical Oncology simultaneous to my talk and is a ten-year culmination of a lot of work by 65 co-investigators that I’ll give tonnes of credit to for assembling guidance of how to construct endpoints in clinical trials that should frame the next generation of new therapies as we apply them in the clinic. Prostate Cancer Working Group 4 defines new nomenclature, new endpoints, particularly using PSMA-PET imaging, which is a novel endpoint for interpreting response as well as progression over time. We integrate that into how patients feel, their liquid biopsy biomarkers, their PSA levels, their patient-reported outcomes, their disease state, the use of androgen pathway modulators such as androgen deprivation therapy or AR pathway inhibitors.
But we also know that in the clinic we’re faced with sometimes having to work outside of those guidelines. Sometimes we see a patient who is responding by PSA but progressing on imaging, those are discordant examples. The Working Group 4 acknowledges those and seeks to report those separately so that we can identify groups of patients who are benefitting and optimise that care of that individual. Certainly we see patients who have symptomatic relief and are clinically benefitting from a therapy even though maybe they have one or two new lesions on bone scintigraphy or PSMA-PET. We would endorse the motivation to treat those new therapies with local therapies like metastasis-directed radiotherapy.
How do definitions of response and progression in clinical trials differ from what clinicians encounter in real-world practice?
So, taking a step back, the working group guidelines have been around for 30 years and they have informed both clinical trials and then the subsequent care of our patients outside of clinical trials once these agents are approved. For example, 20 years ago Prostate Cancer Working Group guidelines identified the bone scan flare phenomenon as a finding on bone scintigraphy where new lesions could appear in the short term during response to hormonal therapy. That’s relatively common, nearly 20-25% of patients who are responding to treatment see new lesions on bone scans and thus the 2+2 rule for bone scans requiring confirmatory scans to document additional new lesions came from that.
The success of that documentation and protocols allowed patients to stay on study longer, pass what was called pseudoprogression, and not to abandon that therapy because of new lesions that reflected more of a healing reaction in response to therapy. We’re now applying that to PSMA-PET which is a novel imaging modality that we don’t know how to use in the clinic in terms of documenting response. Should we be using the standardised uptake value? Should we be counting new lesions? Should we be measuring the whole body SUV or the total tumour burden? And how should we integrate that into other liquid biopsy tests like PSA level so we provide that kind of guidance so that clinicians are not faced with making these decisions on their own?
What are the limitations of current response criteria when applied outside controlled trial settings?
The Working Group 4 guidance around, for example, PSMA-PET progression criteria are our best effort to identify what we think would be meaningful in terms of when patients could be declared progression. We have this rule of five that distinguishes oligometastatic progression and polymetastatic progression. When you see patients with oligometastatic progression we commonly apply a metastasis-directed radiotherapy, we’re reassured that the patient’s prognosis may still be very good, versus when you see systemic progression with multifocal disease progression on imaging, that generally indicates a need for change in systemic therapy. So it reflects a real need for management change. These guidelines need to be validated prospectively in randomised trials before they’re acceptable to regulatory authorities and we’re always willing to accept evidence-based changes as we develop the next ten years of evidence-based guidelines.
How should clinicians interpret atypical responses, such as pseudoprogression or mixed responses in routine care?
Getting confirmation of results is always indicated. For example, PSAs can go up before they go down; they can go down before they go up, that can be seen with circulating tumour cells, cell-free DNA, alkaline phosphatase, pain flares, imaging flares like the 2+2 guidelines. So getting confirmation of something when a result may be discordant from what you’re seeing in the patient in front of you, if the patient is doing well, they’re clinically well but you’re seeing new lesions, that may indicate pseudoprogression. A PSA that’s going up without the patient having symptoms or imaging progression, we would encourage patients and providers to stick with the therapy as it may be benefitting them in the absence of other disease manifestations. So there are a number of situations where that discordant event comes up and I think it’s important to look holistically at all that’s happening to that patient.
What changes are needed to ensure response evaluation frameworks remain relevant in the era of immunotherapy and targeted treatments?
Immunotherapy and targeted therapy have different patterns of response. Targeted therapy tends to be a short-term response, very rapid; immunotherapy tends to be slower and tends to raise the tail at the end of the Kaplan-Meier curve. So we may evaluate those two therapies differently, we may wish to focus on long-term survival or Kaplan-Meier curves that change over time with lesser short-term benefit, greater long-term benefit for those examples. So I think we need to adapt our trial designs to reflect those endpoints and the mechanisms of action of our agents, to collect as much information so that we’re making go/no go decisions in phase II to give our agents the best chance to make it in phase III so that they will help our patients.
Is there anything else you would like to add?
I’d just like to thank the 65 co-authors and the many working group co-authors, including my co-Chair, Michael Morris and Howard Scher, Memorial Sloan Kettering, who really contributed to the last ten years of this evidence-based document to help bring this into the clinic.
