Sonpavde Reviews Urothelial Carcinoma Treatment Highlights from ESMO 2022


Guru P. Sonpavde, MD, provides commentary on several studies in urothelial carcinoma and how data presented at ESMO may affect clinical practice.

Guru P. Sonpavde, MD

Guru P. Sonpavde, MD

Updates in urothelial carcinoma presented at the ESMO Congress 2022 included hypothesis generating outcomes and potentially practice changing results, according to Guru P. Sonpavde, MD.

Sonpavde reviewed data from trials such as EV-103 (NCT03288545), RACE IT (NCT03529890), JAVELIN Bladder 100 (NCT02603432), IMvigor130 (NCT02807636), the UK study BladderPath (ISRCTN 35296862), as well as 2 additional trials from Dana-Farber Cancer Institute. He contextualized how findings are poised to shake up the treatment landscape for patients with urothelial carcinoma if encouraging activity persists.

“We’ll need to wait to see what the regulatory agencies decide but if the [EV-103 trial] regimen is an option in the cisplatin-ineligible setting [and] if it is approved we need to choose…between the JAVELIN [Bladder 100 regimen]of platinum-based chemotherapy followed by maintenance avelumab [Bavencio] or enfortumab vedoin-ejfv [Padcev]/pembrolizumab [Keytruda].”

In an interview with OncLive®, Sonpavde, the GU oncology director at the Advent Health Cancer Institute in Orlando, Florida, provided commentary on several studies and how the data presented at ESMO may affect clinical practice.


In EV-103, one of the cohorts [cohort K] looked at a randomized comparison in cisplatin-ineligible patients with metastatic urothelial carcinoma where the patients were randomized to enfortumab vedotin alone or enfortumab vedotin plus pembrolizumab. This was in the first-line setting and there were approximately 75 patients per arm. This is a randomized phase 2 study and… the response rate with enfortumab vedotin/pembrolizumab was 64.5% [and] the response rate with enfortumab vedotin alone was approximately 45%. The median duration of response with enfortumab vedotin alone was [approximately] 13 months [and] we don’t have the duration of response reached for enfortumab vedotin/pembrolizumab, so we’re waiting on that.1

The toxicity profile was consistent with what’s been seen with enfortumab vedotin and pembrolizumab in the prior phase 1b part of the study, which was an approximately 45 patient study of enfortumab vedotin plus pembrolizumab in the same first-line, cisplatin-ineligible setting where the response rate was approximately 73%.

What we’ve seen so far suggests that the enfortumab vedotin plus pembrolizumab combination in the first-line, cisplatin-ineligible patients is quite active; the data and the response rate, efficacy, and safety look similar in 2 separate phase 1b or phase 2 nonrandom small cohorts, they are still no phase 3 trials.

We believe that this could potentially be practice impacting because this combination probably will go to the regulatory agencies for at least a provisional approval. Of course, this is also a group of patients where first-line atezolizumab [Tecentriq] or pembrolizumab might be an option in select patients where available.


This is a randomized study in patients with localized bladder cancer [which looked at] staging of patients [for] muscle invasive or non–muscle invasive disease who were going for treatment.2

The current staging standard is for patients to undergo in-office cystoscopy and biopsy and following that they undergo a maximal transurethral resection of the bladder tumor called TURBT; [however,] sometimes it takes [some time] for the patient to get from the initial biopsy to TURBT. Investigators in the UK [enrolled] [approximately 140] patients [in] BladderPath. The study could not accrue rapidly and so it was topped a little early.

There was the standard arm of biopsy followed by TURBT, followed by treatment, which could be treatment for muscle-invasive bladder cancer [MIBC] or non–muscle invasive bladder cancer. The experimental arm was an in-office cystoscopy biopsy followed by MRI, not TURBT, and the investigators were trying to look at which arm gets [patients] to definitive treatment quicker. We have these delays between the initial biopsy and the subsequent TURBTs [that] are sometimes a problem and, potentially, [delays] could compromise efficacy, and long-term outcomes.

What the investigators found was that [patients who underwent] biopsy and MRI got to definitive treatment for MIBC much sooner. It was statistically significant; the difference was approximately 90 days [with TURBT] vs 50 days [with MRI]. That is interesting.

One of the things that was not clear from this presentation is how well does the MRI correlate with pathologic stage at cystectomy. All the patients did not go to cystectomy upfront after the MRI, so we don’t know the answer to that question, but it does save time for these patients that sometimes have to wait a long time to get into the operating room for TURBT or cystectomy, [and delays exist for] neoadjuvant chemotherapy.


This is a nonrandomized phase 2 study which combined an immune checkpoint inhibitor, the PD-1 inhibitor [nivolumab (Opdivo)], with radiation in the neoadjuvant setting. Definitive radiation was 50.4 Gy with nivolumab followed by cystectomy.3

[There were] 33 patients in this study, and the combination was safe and feasible and also seemed active because the pathologic complete remission rate was approximately 38%. What we don’t have a good handle on is the long-term outcomes; these data need further evaluation.

There are ongoing studies using a hypofractionated radiation course, rather than this definitive 50.4 Gy course, which if you might recall hypofractionated radiation could be more immunogenic. That is being looked at in combination with immune checkpoint inhibition as a neoadjuvant therapy.

Biomarker Analysis of JAVELIN Bladder 100

The JAVELIN Bladder 100 trial, which was maintenance avelumab following platinum-based chemotherapy in patients who had stable or responding disease, was a positive trial [and] it led to a change in the standard of care. Investigators looked at a chromatin structure model in peripheral blood, and also looked at immune gene expression signature within the tumor tissue. In the peripheral blood, the chromatin signature was [analyzed] in the peripheral blood cells, the white blood cells, and what was found was the lack of a signature called the POU2F2 in peripheral blood. This is derived from the chromatin signature within the nuclei of these white blood cells.4

The absence of the signature in peripheral blood correlated with increased gene expression, immune gene expression in tumor tissue, and correlated with better outcomes with maintenance avelumab. This is hypothesis generating.


[Investigators performed a] a retrospective analysis [of IMvigor130 and] this was a study in first line advanced urothelial carcinoma that compared platinum-based chemotherapy [with or without]atezolizumab, and a third arm of atezolizumab alone. For this retrospective study, investigators looked at the atezolizumab alone arm and the platinum-based chemotherapy plus placebo, [the analysis was focused on] various biomarkers.5

What they found [via] immune-cell or PD-L1 staining in dendritic cells was prognostic for better outcomes for atezolizumab vs gemcitabine/platinum.

The dendritic cell staining of PD-L1 that is most important. This is interesting [and] we’ve always struggled for a biomarker to select patients for first line PD-1 or PD-L1 inhibition and maybe this warrants further validation.

Pan-Tumor Regression Study (Dana-Farber)

[In the first study from Dana-Farber and collaborators] investigators looked at what we call ART, any regression of tumor, for correlation with long-term outcomes in patients with solid tumors receiving immune checkpoint inhibition. There were patients with urothelial carcinoma, but there was also lung cancer, melanoma, gastroesophageal, and other cancers in this.6

We hypothesized that any regression, even 1% or 2%, is more meaningful in the context of an immune checkpoint inhibitor because regressions are more durable with an immune checkpoint inhibitor as opposed to chemotherapy or most targeted agents; regressions and responses are not as durable in those settings.

One impetus for the study was that when we do hypothesis-generating retrospective studies categorizing patients into various response groups is challenging in terms of power to discover new biomarkers. Ideally you want to dichotomize patients into group A and group B, so we have the most power; sometimes you may want to look at outliers if you have a different objective. But using any regression will help us to dichotomize patients into regression vs no regression.

The other impetus was that if you have response categories, you will need central radiology review to get response categories and that is not necessary when you’re using any regression of tumor because any regression of tumor is easily identified but even a nonradiologist.

At the end of the day, any regression of tumor strongly correlated with long-term outcomes and was better than stable disease. Stable disease is a category of response which is very heterogeneous; there are patients who are slowly progressing and slowly regressing, and it’s very challenging to use this group when you do hypothesis-building retrospective studies.

We think that any regression of tumor is a valuable intermediate endpoint, for especially retrospective studies, hypothesis generating studies, to discover biomarkers.

Prognostic Model Using STRONG Data

We looked at a prognostic model combining PD-L1 status with clinical factors [to identify optimal responders] among patients receiving an immune checkpoint inhibitor.7

[Using data from] phase 3b safety study [STRONG; NCT03084471] of fixed dose durvalumab [Imfinzi] given every 4 weeks in patients with metastatic urinary tract cancer—not just urothelial, [but] most of patients had urothelial disease. These are patients progressing post platinum-based chemotherapy and this data set included more than 800 patients.

We looked at PD-L1 status, blood neutrophil lymphocyte ratio, visceral metastasis status, platelet count, and performance status. All these variables at the end of the day, a 4-factor model stood out as being prognostic and had good power to predict both response and survival.

The nice thing about this model is all of these are readily available, [assessing] PD-L1 status is cheap, most [practices] can do it anywhere in the world. It’s a lot cheaper than doing genomic profiling of the tumor [and] takes much less time. It’s very practical and a way to identify patients who may or may not benefit. For example, a patient who has unfavorable neutrophil lymphocyte ratio, which is a bad prognostic factor, patients with reversible metastasis, or low PD-L1, this might not be the patient you want to start with the first line PD-1 inhibitor.

This is something to think about it especially when you have many options in these settings. Should you be giving an immune checkpoint inhibitor or should you be giving other drugs? Enfortumab vedotin is out there in the second- and third-line settings; there are other drugs. I think that this is a model that could be useful in the clinic when you have many options in front of you.


  1. Rosenberg JE, Milowsky M, Ramamurthy C, et al. Study EV-103 Cohort K: antitumor activity of enfortumab vedotin (EV) monotherapy or in combination with pembrolizumab (P) in previously untreated cisplatin-ineligible patients (pts) with locally advanced or metastatic urothelial cancer (la/mUC). Ann Oncol. 2022;33(suppl 7):S1441. doi:10.1016/j.annonc.2022.08.079
  2. James ND, Pirrie S, Liu W, et al. First results from BladderPath: A randomised trial of MRI versus cystoscopic staging for newly diagnosed bladder cancer. Ann Oncol. 2022;33(suppl 7):S1330. doi:10.1016/j.annonc.2022.07.1811
  3. Schmid SC, Schiller K, Lewerich J, et al. RACE IT: a prospective, single arm, multicenter, phase II-trial to assess safety and efficacy of preoperative RAdiation therapy before radical CystEctomy combined with ImmunoTherapy in locally advanced urothelial carcinoma of the bladder (AB 65/18) – first results. Ann Oncol. 2022;33(suppl 7):S1442. doi:10.1016/j.annonc.2022.08.081
  4. Powles TB, Sridhar S, Bellmunt J, et al. Genomic biomarkers in peripheral blood (PB) from patients (pts) enrolled in the JAVELIN Bladder 100 trial of avelumab first-line (1L) maintenance in advanced urothelial carcinoma (aUC). Ann Oncol. 2022;33(suppl 7):S1442. doi:10.1016/j.annonc.2022.08.080
  5. Grande Pulido E, Galsky MD, Van Elzen R, et al. PD-L1 expression on immune cells by SP142 co-localises with dendritic cells and is associated with improved overall survival (OS) with atezolizumab in patients with untreated metastatic urothelial cancer (mUC). Ann Oncol. 2022;33(suppl 7):S1329. doi:10.1016/j.annonc.2022.07.1813
  6. El Zarif T, Pond G, Nassar A, et al. Any regression of tumor (ART) as an intermediate endpoint in patients (pts) treated with immune checkpoint inhibitors (ICI): a pan-cancer analysis. Ann Oncol. 2022;33(suppl 7):S591-S592. doi:10.1016/j.annonc.2022.07.148
  7. Sonpavde GP, Sternberg CN, Lee JL, et al. 4-factor model to predict response and survival benefit with durvalumab (D) in previously treated metastatic urinary tract carcinoma (mUTC). Ann Oncol. 2022;33(suppl 7):S1337. doi:10.1016/j.annonc.2022.07.1826
Related Videos
Vikram M. Narayan, MD, assistant professor, Department of Urology, Emory University School of Medicine, Winship Cancer Institute; director, Urologic Oncology, Grady Memorial Hospital
Nikhil Gopal, MD, assistant professor, urology, College of Medicine, Memphis Department of Urology, The University of Tennessee Health Science Center
Jonathan E. Rosenberg, MD
Joaquim Bellmunt, MD, PhD
Arlene O. Siefker-Radtke, MD
Alexander Kutikov, MD, FACS, of Fox Chase Cancer Center
Roger Li, MD, of Moffitt Cancer Center
Joshua J. Meeks, MD, PhD, of Northwestern University Feinberg School of Medicine
Neal Shore, MD, FACS, of GenesisCare USA and Carolina Urologic Research Center
Mark D. Tyson, II, MD, MPH