Immuno-Oncology Resource Center
Immuno-Oncology Resource Center
July 25, 2019
5 min read

A new love affair — but will you respect me in the morning?

Despite 'real progress' in bladder cancer treatment, cure rate with immunotherapy remains to be defined.

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

We have known for more than 50 years that immunotherapy is active against noninvasive bladder cancer, dating from the paradigm-altering work of Donald Lamm, MD, FACS, Alvaro Morales, MD, the EORTC investigators and others.

Innumerable single-arm and randomized trials have confirmed the utility of the various strains of bacillus Calmette-Guérin, or BCG, in treatment and prevention of recurrence of noninvasive disease. Interferon also has demonstrated some activity in this context.

Derek Raghavan, MD, PhD, FACP, FRACP, FASCO
Derek Raghavan

Most randomized trials have shown that intravesical administration of BCG is more effective than intravesical cytotoxic chemotherapy, although it is clear that mitomycin and doxorubicin, both with large molecular weights, have anticancer efficacy against noninvasive bladder cancer without excessive systemic toxicity (because these large compounds generally are not absorbed through an intact bladder wall).

Thirty-five years ago, a very small series from Netto and Lemos, working in Brazil, reported some anticancer efficacy of oral BCG against invasive bladder cancer. Their data were not confirmed by subsequent, equally small studies, and this approach to invasive disease has lain fallow for decades.

Renaissance of IO

With the recent renaissance of immuno-oncologic (IO) agents, there has been progress in the management of advanced and metastatic bladder cancer for the first time in about 20 years.

Despite a wealth of second-line cytotoxic trials, and the several variants of the COXEN study, it has been clear that cure or even long-term survival is very unlikely in response to second-line cytotoxic treatment once MVAC (methotrexate, vinblastine, doxorubicin and cisplatin) or gemcitabine-cisplatin (or equivalent) regimens have failed. Most responses last only a few months, and despite breathless P value and waterfall or forest plot citations, the reality has been that early death from relapsed metastatic bladder cancer is inevitable.

With the advent of IOs targeting PD-L1 or CTLA-4 inhibition, many of these agents — such as pembrolizumab (Keytruda, Merck), nivolumab (Opdivo, Bristol-Myers Squibb), atezolizumab (Tecentriq, Genentech) and others — have demonstrated quite substantial activity after the failure of cisplatin-based regimens, with objective response rates of around 20% to 30% and median survival of about 9 to 10 months.

The initial modeling was done in “cisplatin-ineligible” patients, although it was never very clear whether that term reflected patients who were truly ineligible or who simply were defined by inadequate workup with failure to exclude tumor-based outflow obstruction as the cause of cisplatin-limiting renal dysfunction. Obviously, the correct approach in the setting of tumor-based renal dysfunction in those cases would have been to insert a stent or a nephrostomy, wait 1 to 2 weeks for renal function to improve, and then use a cisplatin-based frontline regimen.


It is now clear that these agents have activity in the first line, irrespective of renal function, based initially on phase 1 to phase 2 studies.

CheckMate 032

An important study was reported in this month’s Journal of Clinical Oncology, which I think ultimately could be paradigm-shifting (with caveats) and thus worthy of discussion here.

In CheckMate 032, an international expansion cohort study supported by Bristol-Myers Squibb, patients with relapsed urothelial cancer after cisplatin-based therapy received either 3 mg/kg nivolumab once every 2 weeks; 3 mg/kg nivolumab plus 1 mg/kg ipilimumab (Yervoy, Bristol-Myers Squibb) every 3 weeks for four doses followed by 3 mg/kg nivolumab monotherapy every 2 weeks (N3-I1 regimen); or 1 mg/kg nivolumab plus 3 mg/kg ipilimumab every 3 weeks for four doses followed by 3 mg/kg nivolumab monotherapy every 2 weeks (N1-I3 regimen).

I have no idea who figured out that complex and not completely logical trial design. Nonetheless, despite rather small numbers accrued to each arm of the study (three arms with less than 400 patients overall), the overall response and survival figures are quite impressive, given that this represents platinum-resistant disease.

Overall response rates attributed by a blinded central review panel were 20.5% (95% CI, 12.2-31.2) for nivolumab monotherapy and 37% (95% CI, 27.1-47.7) for N1-I3. Central review data were not reported for the N3-I1 regimen, but investigators attributed an ORR of 26.9% (95% CI 18.7-36.5).

Although perhaps drawn to emphasize the differences, in absolute terms the waterfall plots suggested that the N1-I3 regimen conferred more substantial responses.

Median OS was 9.9 months for nivolumab monotherapy, 7.4 months for the N3-I1 regimen and 15.3 months for the N1-I3 regimen, quite impressive for this context, especially as more than 80% of each group had visceral metastases at entry.

Of importance, the survival curve for N1-I3 was superior throughout and, intriguingly, the survival tail beyond 2 years, although underpowered, showed several long-term survivors. Further follow-up and independent, nonpharma-supported confirmation of the utility of this regimen may well lead to a fundamental practice change, with even more support for the role of IOs for advanced bladder cancer.

Not to be forgotten

However, let us have a brief reality check.

In my quaternary referral practice, I have had to advise some oncologists, who have used multiple IOs for advanced bladder cancer and then requested advice, that cytotoxic cisplatin-based chemotherapy remains active for urothelial cancer and should not be forgotten.

As is so often the case, modern oncology has a love affair with anything new, forgetting that the full natural history of novel therapies has not yet been charted, that patterns of late toxicity have not yet been fully defined, and that the length of time for measurement of treatment failure is usually much longer for established treatments than for novel therapies (and thus biases the comparison of the two).


My point is simple. There is an appropriate and effective revolution in bladder cancer management associated with the introduction of IOs, and this is real progress. However, the percentage of patients achieving permanent cure has not yet been defined.

The pattern of toxicity — particularly in the hands of inexperienced oncologists — can be very substantial, especially as the early toxic changes are often very subtle and can be missed at a time when management may be easier and more effective.

Finally, we should not forget that the cisplatin-based cytotoxic regimens are associated with cure for patients with distant node metastases, and even for some patients with biopsy-proven lung metastases.

After decades of stalled progress, bladder cancer therapy is moving forward with real and important steps that are paradigm-shifting.

In introducing them into our routine therapy algorithms, we must not forget the utility of cytotoxic agents in this disease, and we must be mindful of the fiscal and physical costs of treatment with IOs and of the importance of experience of the physicians who use them. In addition, the nuances of optimal patient selection have yet to be defined.

The question remains, after the love affair today, will there be respect tomorrow morning?


Netto NR Jr and Lemons GC. J Urol. 1984;132:675-677.

Sharma P, et al. J Clin Oncol. 2019;doi:10.1200/JCO.19.00538.

For more information:

Derek Raghavan, MD, PhD, FACP, FRACP, FASCO, is HemOnc Today’s Chief Medical Editor for Oncology. He also is president of Levine Cancer Institute at Atrium Health. He can be reached at

Disclosure: Raghavan reports no relevant financial disclosures.