Dual ICI Considered for a Patient With NSCLC and Low PD-L1
During a Case-Based Roundtable® event, John V. Heymach, MD, PhD, discusses the use of immunotherapy combinations based on patients PD-L1 expression in the first article of a 2-part series.
John V. Heymach, MD, PhD, summarizes case for a 62-year-old patient with non–small cell lung cancer.
John V. Heymach, MD, PhD
Professor, Department of Cancer Biology
David Bruton, Jr Chair in Cancer Research
Chair, Department of Thoracic/Head and Neck Medical Oncology
Division of Cancer Medicine
The University of Texas MD Anderson Cancer Center
Houston, TX
Targeted Oncology: Does this patient’s low PD-L1 expression impact your treatment choice?
JOHN V. HEYMACH, MD, PHD: When your patient has negative PD-L1 expression, the duration of response [DOR] to PD-1 inhibitor monotherapy tends to be much shorter. That's part of the reason I tend to go towards dual immunotherapy at some point in patients with the PD-L1 expression less than 1%. But you certainly could use monotherapy. The KEYNOTE-407 [NCT02775435] is a perfectly legitimate regimen [in this case]. It's FDA approved, and since we don't usually use maintenance chemotherapy with squamous disease, you could give the 4 cycles of chemotherapy with pembrolizumab [Keytruda] and then keep the pembrolizumab going. That is a legitimate data-supported choice.
What other regimens are viable for a patient such as this with non–small cell lung cancer (NSCLC)?
CheckMate 9LA [NCT03215706] was with ipilimumab [Yervoy]/nivolumab [Opdivo], [investigating] dual immune checkpoint inhibitor [ICI] with chemotherapy vs chemotherapy alone.1,2 The experimental arm has only 2 cycles of chemotherapy, but then you keep the ipilimumab/nivolumab going in the maintenance setting for as long as it's tolerated. This is the spread-out dosage of ipilimumab, 1 mg/kg every 6 weeks, as opposed to the melanoma dosing with 3 mg/kg every 3 weeks.... I tend to be pretty quick to drop the ipilimumab if patients are having immune toxicities.
What was the efficacy of ipilimumab/nivolumab seen in the 5-year update of CheckMate 9LA?
For patients with NSCLC [in the overall population], the overall survival [OS] HR was 0.73 [95% CI, 0.62-0.85].2 The median OS was 15.8 vs 11.0 months [for ICI/chemotherapy vs chemotherapy, respectively]. So it's adding 5 months or so in terms of the median OS for all patients. We know that patients with PD-L1 expression less than 1% [usually] do worse than those with greater than 1% [in terms of survival]. But if you look at the difference in HRs, you gain a lot more from this regimen in CheckMate 9LA. The HR was 0.63 [95% CI, 0.49-0.83] in the PD-L1 less than 1% group as compared with the PD-L1 greater than 1% group where the HR was 0.73 [95% CI, 0.59-0.90] for OS. In the absolute amount gained, patients gain about 8 months in the PD-L1 less than 1% arm and 4 to 5 months in the PD-L1 greater than 1% arm. We know that the PD-L1 less than 1% population does worse, but they seem to get more relevant benefit from the CheckMate 9LA regimen as compared with chemotherapy.
At 5 years, this benefit seems to be holding up. I'm impressed if you look at the patients with squamous NSCLC at 5 years, 18% vs 7% OS rate [with ipilimumab/nivolumab/chemotherapy vs chemotherapy], so more than doubling the likelihood of being alive at 5 years. If you look at the patients with nonsquamous NSCLC, it was 19% vs 12%, respectively. Still worthwhile, but not nearly the same magnitude of benefit.
The DOR, if you look at PD-L1 less than 1% population, that median DOR was pretty dramatic with the experimental arm at 17.5 months vs 4.3 months with chemotherapy alone. If patients respond to this regimen, that median DOR is impressive for these patients. For patients with PD-L1 expression greater than 1%, for reasons I don't understand, the median DOR was shorter at 11.8 months with ICI/chemotherapy compared with chemotherapy at 5.6 months. This could be because this population may have a higher tumor mutation burden—I'm not sure what it is. But there was an impressive DOR for the PD-L1 less than 1% population.
The OS HR...for PD-L1 expression less than 1% and the squamous populations seem to be the groups that get particularly large benefit from this regimen. Those are the groups that tend to benefit less from the pembrolizumab combinations.
How did patients do in terms of toxicity on the CheckMate 9LA trial?
The immune-related adverse events [AEs] for the ipilimumab/nivolumab/chemotherapy arm, in the first 1 to 5 doses of ipilimumab...the rates of grade 3/4 [toxicity included] 17% rash and 25% hepatitis. I've also seen a decent amount of colitis, although the rates that they're reporting here are a little bit lower; the colitis tends to kick in from between doses 6 and 15. So patients get rash and hepatitis early, and the colitis later in terms of the toxicity. The hypothyroidism and hyperthyroidism tend to kick in later as well. But for grade 3/4 AEs, rates are relatively low. In my experience, the diarrhea, colitis, and hepatitis tend to be manageable with just steroids and holding the dose. There was 1 case of grade 3/4 adrenal insufficiency in doses 1 to 5 and 6 to 15 each.
