Skip to main content


Pembrolizumab Unlikely Cost-Effective for BCG-Unresponsive Bladder Cancer in US

February 18, 2021


Vidit Sharma, MD, University of California, Los Angeles, Department of Urology, discusses results from a cost-effectiveness analysis comparing pembrolizumab with radical cystectomy or salvage intravesical chemotherapy for patients with BCG-unresponsive carcinoma in situ of the bladder.

These results were presented at the virtual 2021 ASCO Genitourinary Cancers Symposium.


Hello, everyone. I'm Vidit Sharma. I'm a health services research fellow in urologic oncology and also a urologic oncology fellow at Mayo Clinic, and I'm doing my health services research fellowship at UCLA.

I'll be presenting our study titled, "The Cost Effectiveness Analysis of Pembrolizumab for BCG-Unresponsive Carcinoma In Situ of the Bladder."

Just some background, standard of care for BCG-unresponsive carcinoma in situ is radical cystectomy and salvage intravascular chemotherapy for appropriate patients.

Recently, the FDA has approved pembrolizumab for BCG-unresponsive CIS on the basis of a single‑arm trial, the KEYNOTE 57 trial, which demonstrated a 3‑month response rate of 41% and a 2‑year recurrence‑free survival of about 15%.

In light of this response rate, there's also potential downsides of pembrolizumab, one being that it costs over $100,000 per year if a patient continues it for the full year, and it also has a toxicity rate of over 60%. It's unclear if the benefits of pembrolizumab are sufficient to warrant its widespread use in light of its known harms.

Our objective was to conduct a cost-effectiveness analysis for BCG-unresponsive CIS, comparing pembrolizumab, radical cystectomy, and salvage intravascular chemotherapy. For our salvage intravascular chemotherapy, we use gemcitabine-docetaxel as our prototypical regimen.

In our analysis, we compared these 3 options for 2 index patients. Index patient 1 was willing and able to undergo a radical cystectomy. Index patient 2 was either unwilling or unable to undergo a radical cystectomy.

We used the decision analytic Markov model with several health states, each of which were unique for each of these treatment options. Overall, our model had in excess of 100 inputs and parameters. We tried to be as thorough as we could to mimic the clinical practice with these agents and treatments.

Lastly, we also performed several sensitivity analyses in which we varied the model inputs to see how the conclusions of the models changed as we changed the input values of the model.

The key finding was that the 5‑year cost of radical cystectomy and salvage intravascular chemotherapy were $39,000 and $43,000 respectively, but pembrolizumab actually cost an excess of $191,000 at 5 years.

This was accounting for the fact that patients who had pembrolizumab and had a subsequent recurrence stopped pembrolizumab at the time of the recurrence. The cost of pembrolizumab was only up to the time when pembrolizumab would have been given in the trial.

When we examined the quality‑adjusted life years, we found that pembrolizumab only added about 1 quality‑adjusted life month, or about 0.1 quality‑adjusted life year over radical cystectomy, despite the 4‑fold increase in cost.

We found that pembrolizumab was not cost effective relative to radical cystectomy or salvage intravascular chemotherapy for either index patient 1 or 2, which was related to their cystectomy eligibility. This is reflected by incremental cost-effectiveness ratios of over $1 million when you compare pembrolizumab to either radical cystectomy or salvage intravascular chemotherapy.

Then I just want to add that we did have somewhat optimistic assumptions in favor of pembrolizumab, and we tried varying several model parameters in our sensitivity analysis. Really, the major parameter that did convert pembrolizumab to being cost-effective was a price reduction of over 90%.

Salvage intravascular chemotherapy was close to being cost-effective, relative to radical cystectomy, with an incremental cost-effectiveness ratio of about $118,000, and the threshold that we used for cost-effectiveness was $100,000 per quality‑adjusted life year.

In our sensitivity analyses, salvage intravascular chemotherapy actually became cost-effective if certain recurrence and metastasis thresholds were met in our model. Using the data that we extracted from the literature, those thresholds were actually very close to the data in the literature. They were just slightly below.

We conclude that our model supports the preferential use of radical cystectomy or gemcitabine-docetaxel over pembrolizumab, and further prospective studies of gemcitabine-docetaxel are warranted to determine if it meets our model threshold and it's a cost-effective option for patients in this space.

We also hope that this model can become a platform to evaluate single‑arm trials for other agents in the BCG‑unresponsive space. Thank you very much for your attention.

Sharma V, Wymer K, Saigal C. Cost-effectiveness analysis of pembrolizumab for BCG-unresponsive carcinoma in situ of the bladder. Presented at: the virtual 2021 ASCO Genitourinary Cancers Symposium; February 11-13, 2021. Abstract 395. 

Dr Sharma is a VA Health Services Research & Development Fellow. He reports no additional relevant financial relationships.

Back to Top