The Role of Imaging in Neoadjuvant Immunotherapy in Muscle-Invasive Bladder Cancer: A Case Study of PURE-01 - Andrea Necchi & Laura Marandino

May 11, 2021

Alicia Morgans talks with Andrea Necchi and Laura Marandino. They delve into the PURE-01 study, which focuses on the use of neoadjuvant pembrolizumab in muscle-invasive bladder cancer patients. Dr. Necchi elaborates on the study's primary and secondary endpoints and highlights the role of PET-CT imaging in assessing treatment response. The study finds that PET-CT has limited utility in predicting pathological lymph node involvement, but it may help identify poor responders to treatment. Dr. Marandino discusses her work on detecting immune-related adverse events through PET-CT, revealing that 39% of patients developed such events, most commonly in the thyroid and stomach. Both experts agree that while PET-CT has limitations, it holds potential for clinical research, particularly when combined with other biomarkers. The conversation underscores the need for further studies to refine patient selection and treatment approaches.


Andrea Necchi, MD, Medical Oncologist, Department of Medical Oncology, Fondazione IRCCS - Istituto Nazionale dei Tumori, Milan, Italy

Laura Marandino, MD, Medical Oncologist, Research Fellow, Oncologist Institute of Southern Switzerland, Bellinzona, Switzerland

Alicia Morgans, MD, MPH Genitourinary Medical Oncologist, Medical Director of Survivorship Program at Dana-Farber Cancer Institute, Boston, Massachusetts.

Read the Full Video Transcript

Alicia Morgans: Hi, my name is Alicia Morgans. I'm a GU Medical Oncologist and Associate Professor of Medicine at Northwestern University in Chicago, in the United States. I'm so excited to have here with me today, Dr. Andrea Necchi, who is the head of GU Medical Oncology at San Raffaele Hospital in Milan, Italy.  As well as Laura Marandino, who is a fellow at the Oncology Institute of Southern Switzerland and who has worked with Dr. Necchi and some of his really exciting work looking at imaging, particularly PET-CT imaging in patients in the PURE-01 study. Thank you both so much for being here with me today.

Andrea Necchi: Thank you, Alicia. It's a pleasure to be here with Laura and the UroToday people, and yourself to discuss, again, issues around the PURE-01 study, thank you.

Laura Marandino: Thank you, Alicia.

Alicia Morgans: Thank you, Laura. So, why don't we start with you Andrea, just reminding us what the PURE-01 study is and then telling us about this first study, where you really investigated the use of PET-CT in these patients, who are getting neoadjuvant therapy before going to cystectomy.

Andrea Necchi: Thank you for the opportunity to further highlight our work, Alicia. So the PURE-01 study is a pretty well-known study in muscle-invasive bladder cancer, in which we provide the three cycles of neoadjuvant pembrolizumab in clinically N0 patients, and before radical cystectomy. The primary endpoint of the PURE-01 study was the pathologic response and the secondary endpoints were survival outcomes. Additional endpoints, translational endpoints have been represented by imaging endpoints. We debated a lot regarding biomarker tissue, related biomarker data from this study. But the one thing that we further highlighted is the role of imaging, for advanced imaging to stage the tumor and to reassess the response of treatment to immunotherapy. So, applying these tools in a relatively new context of neoadjuvant immunotherapy.  The cohort to PURE-01 included 146 patients, of which more than 100, 108 in particular patients, that had matched pre and post-therapy, PET-CT scans.

And so were available for analysis on PET-CT, and they investigated the role of PET-CT as well as other imaging that were previously reported like a multiparametric MRI of the bladder. But now focusing on the PET-CT scan, FDG PET-CT scan, we try to make some assessment that, by comparing the images at baseline and relating the images to the pathological outcome data and to the clinical outcome data. So, the first observation that we make in relation to the pathologic response and today with the aim to predict somehow, and with the PET-CT images that the response to the pembrolizumab and the assessment of the pathological status of the other lymph nodes of radical cystectomies, we found that actually, that this tool, so PET-CT applied to already selected patients with a clinical N0 patient, based on conventional CT scan, did not get substantial information regarding the possibility to predict the pathological involvement of lymph nodes.

Because actually what we found was that the sensitivity was quite low, was around 27%, 25%. That's in line pretty much in line with the few data available in the literature, conversely, the specificity is pretty high, was around 90%, 95%. The accuracy is intermediate, was something around 60% and 70%. But overall, the message related to the potential application of this tool, as a further staging any routine clinical practices that PET-CT is limited in terms of the possibility to further assess or predict somehow the response to treatment in the pathological lymph nodes. Besides this, what we found is that the very few patients that we found to have an FDG uptake at baseline, despite having a negative CT scan, so lymph nodes with short access, less than one centimeter, the majority of these patients, 57% of these patients were non-responders to treatment compared to 10% of the rest of the cohort.

Meaning that the very few patients that were found to have some signal of FDG uptake at baseline, despite an N0 status were poor responders. Meaning that this tool is not completely bad in this setting.  Further research perhaps by investigating additional tracers in this disease may further provide us with some hints towards the possibility for the select patients that are more suited for a nonchemotherapy approach, an experimental therapy approach given in the neoadjuvant stage. My impression is that overall, this is still something that cannot be applied in routine practice, but it is still something that is well-suited for clinical research in this field. So this is the main finding related to the association between the PET-CT outcome, PET-CT images, and pathological outcome data.

Then the main message is that the PET-CT scan actually is a tool that should be confined to clinical research. And then we have two large cohorts, they are the only cohorts that are reported in the literature so far related to clinical N0 patients with the use of PET-CT in the data presented by our cohort and by the cohort reported by people and colleagues from Memorial Sloan Kettering Cancer Center. Two very similar cohorts, for one side, chemotherapy-treated patients or radical cystectomy only treated patients. On the other side, immunotherapy treated patients, but overall leading to the same message of the sensitivity of this tool, that the predictive with other logical outcome data, but still, we may have some possibility to select the patients. If very few patients with some FDG uptake at baseline in candidates is two non-chemotherapy approaches, neoadjuvant.

Alicia Morgans: Thank you so much. I think it is so critical to really drive this point home because patients ask here in the US all the time, why don't I get a PET scan with my staging for my urothelial cancer? Everybody tells me that PET scans are the best way to really understand cancer. And it is very clear that with such poor sensitivity in this cohort, which is a nicely sized cohort, and all of these patients went to cystectomy, that with a sensitivity that is so low, this is not adding something in terms of that management preoperatively and understanding whether these patients will have positive lymph nodes and that is something that we can emphasize to our patients in an updated cohort and so important. It's incredibly interesting to find though, that those patients who did seem to have positive lymph nodes on the PET imaging were actually the poorest responders to the neoadjuvant pembrolizumab approach, which is a little bit counter-intuitive, but really interesting.

So could we perhaps use this to identify patients who should use chemotherapy in their upfront approach to neoadjuvant treatment? Perhaps! And I look forward to you really helping us understand that in the future with the ongoing work that you and your team are doing. And let's turn to Laura for a moment. Now, Laura, you looked at these PET images to try to understand whether we could predict or understand immune-related adverse events better in this patient population. What did you find?

Laura Marandino: So we checked on those patients who had a match pre and post-therapy plus pembrolizumab PET-CT, the ability of PET-CT to detect immune-related adverse events. So 103 patients were included in this analysis and a PET-CT was considered suggestive of immune-related adverse events, whenever a new onset of inflammatory uptake outside of the physiologic carriers was found compared to the baseline. And we found that 40 patients, in total 39% of patients developed immune-related, adverse events to PET-CT, and the most frequently affected site was the thyroid, which was the most frequently affected organ in 18 patients followed also by the stomach in 14 patients, additional lymph nodes in nine patients, and lung in five patients.

And Interestingly, less than half of these patients had clinical associated adverse events, most frequently thyroid, so most frequently hyperthyroidism, and a lot of patients still were completely symptomatic. And we also checked for the association of the development of new related adverse events and clinical outcomes. And we found that even if there were higher rates of pathologic complete response or downstaging, also relapse-free survival at 24 months, there was not a statistically significant difference. So we have to consider that these are post-hoc exploratory analyses.

Alicia Morgans: I agree, it's important to remember that these are post-hoc exploratory analyses but, if the development of these immune-related adverse events radiographically may be correlated with outcome, that can be helpful. And I thought it was so interesting that a number of these patients didn't have clinically manifested adverse events, but certainly had this radiographic adverse event, particularly since the second most common event was something in the stomach, which is not something that I typically see patients manifest. I'd love to hear your thoughts or yours, Andrea, on these adverse events. Are they meaningful if they are not clinically meaningful? Well, perhaps if they associate with the outcome that's interesting and important. And what is this gastric involvement with this adverse event, what is that?

Laura Marandino: All of our patients were completely asymptomatic for adverse events to the stomach, and also those patients with increased uptake in the mediastinal lymph nodes were completely asymptomatic. And I think it's interesting that for example, also two studies of neoadjuvant immunotherapy in lung cancer reported the same percentage of patients with increased uptake in the hilar and mediastinal lymph nodes that revealed to be granuloma at the examination. So all these data in the neoadjuvant setting until now, there are few but are in accordance to each other.

Alicia Morgans: Very interesting, what is your take, Andrea?

Andrea Necchi: I think that overall it's nice work, but I think that we are still observing the tip of the iceberg because we still need longer follow-up time. Because I think that we are realizing the PURE-01 study in general, that there was a lot of moderated adverse events months after radical cystectomy and cases of a variety of reactions, rash, fevers, post-operative fevers, a lot of possible adverse events that may be somehow predicted wherewith the imaging just after or before intervention. And also another point is that there is an association potentially with surgical safe events because in Steinel uptakes, for example, may somehow predict response.

So an additional focus of course is needed in patients treated with some kind of immunotherapy or even with combination immunotherapy in the next future preoperative, of course. So we need more time, we need more data, in particular data on the surgical safety of these new approaches, preoperatively, of course.

We are still at the beginning of the journey and we will see if larger data, we have a larger phase three trial that is almost close to being concluded and presented in the same setting, will be providing additional information regarding these, these are important aspects about the general management of the patients. And a final consideration may be related to the potential association with the circulating inflammatory biomarkers, and this is an issue that is still the focus for the research. So it's important also to somehow corroborate our imaging data with the circulating biomarker data related to the cytokine. So rather circulating inflammatory biomarker that may somehow be a further predictor of the development or the anticipation of the development of important safety events in patients undergoing immunotherapy. This may also apply, not only to the massive invasive clinical state, but also in general to patients undergoing immune checkpoint inhibitors, of course.

Alicia Morgans: Well, Laura, I think that you, even if you are now in Switzerland, have a lot of work to do, projects laid out by Andrea. And certainly you, Andrea have more work to share with us. We look forward to hearing these updated data, if you can get them with longer-term follow-up in terms of the surgical safety and the other biomarker assessments that will help inform us as we try to understand better, which patients will respond, which patients won't respond and how we best use these imaging modalities to understand both adverse events and perhaps choose patients or use them in order to correctly, pair patients with a proper neoadjuvant treatment approach, which may be in our future as well. So thank you both so much for your time, for your ongoing work. So much more to do, but little steps, step-by-step we will make our way, and thank you for your time and your expertise.

Andrea Necchi: Thank you, Alicia, and UroToday people, it's always a pleasure and has been a pleasure to further discuss this with you, thank you.

Alicia Morgans: Thank you very much Alicia, and thanks to all of you, of course.