PracticeUpdate Oncology May 2019

EDITOR’S PICKS 17

standard doses for single agents (ipi 1 mg/ kg, nivo 3 mg/kg). The primary endpoint of the study was to determine if the ipi-1 reg- imen was better tolerated than the ipi-3 regimen, as measured by grade 3–5 tox- icity. Significantly, the authors found that the lower dose of ipilimumab combined with 3 mg/kg of nivolumab was less toxic in the primary endpoint (grade 3–5 tox- icity of 34% in ipi-1 mg/kg arm vs 48% in ipi-3 mg/kg arm, P = .006) and in all other toxicity measurements, including serious adverse events and adverse events lead- ing to discontinuation. The authors also performed an underpow- ered descriptive analysis of efficacy. The two regimens did not appear to differ mark- edly in secondary efficacy endpoints. The ipi-3 arm had a slightly numerically higher objective response rate than the ipi-1 arm (50.6% vs 45.6%), but median PFS and 12-month OS were virtually indistinguish- able between the two arms. The authors point out that the trial was not designed to establish noninferiority and that there- fore no claims can be made; but, clearly, the overlapping survival curves raise the possibility that the substantial decrease in toxicity may not be accompanied by a cor- responding decrease in efficacy. Patients in the low-dose ipilimumab arm were more likely to complete therapy than those in the standard-dose ipilimumab arm, and were less likely to discontinue therapy due to toxicity, which could account for the sug- gestion of equivalent efficacy despite lower dosing of ipilimumab. Three important caveats should be men- tioned. First, although the toxicity of low-dose ipilimumab + standard-dose nivolumab is better than standard-dose ipilimumab + low-dose nivolumab, it is very important to stress that noninferiority

Despite these caveats, the low-dose combined ipilimumab regimen may be an important "middle-ground" for oncologists and patients who want to achieve potential benefit of combination therapy without the very high rate of toxicity seen in standard- dose ipilimumab combinations. The suggestion of noninferiority in the informal efficacy analyses with clearly improved toxicity in the ipi-1/nivo-3 arm may well assist low-dose ipilimumab combination therapy in replacing the ipi-3/nivo-1 standard regimen that is currently used. However, the currently available data do not in any way support the substitution of single-agent PD-1 inhibitor therapy with this new dosing of combination therapy. The decision of whether to use PD-1 alone or in combination should not be influenced by this study. We must hope for further, better-designed trials to clarify whether de-escalation strategies can truly maintain efficacy while reducing toxicity. References 1. Larkin J, Chiarion Sileni V, Gonzalez R, et al. Combined nivolumab and ipilimumab or monotherapy in untreated melanoma. N Engl J Med 2015;373(1):23-34. 2. Postow MA, Chesney J, Pavlick AC, et al. Nivolumab and ipilimumab versus ipilimumab in untreated melanoma. N Engl J Med 2015;372(21):2006-2017. 3. Tawbi HA, Forsyth PA, Algazi A, et al. Combined

of efficacy has not been demonstrated. This is a serious oversight on the part of investigators, as the suggestion of virtual equivalence in efficacy cannot be acted upon without a formal noninferiority analy- sis. We therefore need to assume that the decrease in toxicity with the low-dose ipili- mumab regimen may well be accompanied by a corresponding decrease in efficacy. Second, single-agent PD-1 inhibitor therapy was not tested in this study. As mentioned previously, single-agent nivolumab or pem- brolizumab shows clearly improved toxicity as compared with the ipi-3 regimen, and has also not been proven inferior with regard to efficacy. Although the toxicity of the low-dose ipilimumab combination reg- imen may be better than standard-dose ipilimumab combination, it still does not come close to the low toxicity rate seen with PD-1 inhibitors alone. As such, we now have three potential regimens available for first-line melanoma, including single-agent PD-1 inhibitors, low-dose ipilimumab com- bination therapy, and standard-dose ipilimumab combination therapy. None of these regimens has definitively shown that it is more effective than any of the others. However, single-agent PD-1 inhibitors are clearly less toxic than standard dose-ipili- mumab combination therapy, and are likely less toxic than low-dose ipilimumab com- bination therapy. And, third, patients in this study could not have active brain metastases. We know that standard-dose ipilimumab combined with low-dose nivolumab is very effective in melanoma with active brain metasta- ses, 3 and, therefore, no conclusions about the feasibility of reducing the ipilimumab dose in this population of patients should be made.

nivolumab and ipilimumab in melanoma metastatic to the brain. N Engl J Med 2018;379(8):722-730.

Dr. VanderWalde is the Director of Clinical Research at the West Cancer Center and Research Institute, as well as Associate Vice Chancellor of Research and Associate Professor of Hematology/Oncology at the

University of Tennessee Health Science Centre in Memphis, Tennessee.

VOL. 3 • NO. 2 • 2019