Future development of novel immunotherapy combinations should be based on individual patient characteristics, such as patient-level immune targets and tumor microenvironment.
Jason J. Luke, MD
Future development of novel immunotherapy combinations should be based on individual patient characteristics, such as patient-level immune targets and tumor microenvironment, according to a presentation by Jason Luke, MD, at the 36th Annual CFS®.
As single-agents, the immune checkpoint inhibitors have demonstrated utility across a broad spectrum of tumors. Additionally, some combinations of anti—PD-1 and anti–CTLA-4 inhibitors have gained approval along with immunotherapy and chemotherapy combinations. However, the recent failed phase III trial looking at pembrolizumab (Keytruda) and epacadostat has called research practices into question.
“The field has moved to a place where we need to be a little bit more dogmatic about things,” said Luke, assistant professor of Medicine at the University of Chicago Medicine & Biological Sciences. “The gold standard still remains randomized trials, whether that be phase II or III trials. We also need real biomarkers, I don't mean PD-L1, I mean hard biomarkers, like BRAF mutations. Without that, it's hard to know who will benefit from the therapy.”
The creation of rational combinations hinges on the ability to uncover an effective biomarker for immunotherapy. To this end, Luke suggested that phenotyping tumor microenvironments into T cell inflamed (also known as “hot” tumors) and non—T cell inflamed (so-called “cold” tumors) at baseline could help guide treatment selection. Distinct genes have been tied to each of these immune states, making baseline screening feasible.
The phase I ADVISE trial is currently testing the plausibility of this strategy across several types of solid tumors. Patients entering the trial will be screened for biomarkers and matched to a corresponding immunotherapy combination. Each group will receive nivolumab (Opdivo) plus a biomarker defined investigational treatment, including relatlimab for LAG3, BMS-986205 for IDO, cabiralizumab for CSF1R, ipilimumab (Yervoy) for FOXP3, BMS-986156 for GITR, and lirilumab for KIR. Additionally, a final arm is looking at nivolumab plus stereotactic body radiation therapy. The study plans to recruit 50 patients, with an estimated primary completion date in January 2020 (NCT03335540).
Another approach for non-T cell inflamed tumors looks to improve the number of T cells in circulation, essentially turning a “cold” tumor “hot.” IL-2 is a proven approach for this and has shown efficacy for renal cell carcinoma and melanoma. In early phase trials, the pegylated CD122 (IL-2R beta gamma) agonist NKTR-214 showed some synergy with PD-1 inhibition.
“The concept here is that targeting the [IL-2] beta gamma subunit will disproportionately have an impact for IL-2 on CD8 cells relative to CD4 Tregs," Luke said. "It was always thought that one of the reasons IL-2 was not more effective is that you stimulated the regulatory mechanisms as much as you stimulated the CD8 mechanisms."
In a phase II dose expansion trial,1 the overall response rate (ORR) was 50% for the combination of nivolumab plus NKTR-214 in treatment-naïve patients with melanoma. In those with PD-L1—negative disease, suggesting a “cold” tumor, the ORR was 42%. An open-label phase III study is currently exploring NKTR-214 plus nivolumab for untreated patients with metastatic melanoma (NCT03635983).
“This is early data from the clinical trial, and there's some really interesting things going on here,” Luke said. “A number of patients who supposedly have PD-L1—negative tumors are having really deep responses.”
In addition to these strategies, site-specific immune inflammation has been examined using intralesional injections of oncolytic viruses. The oncolytic virus talimogene laherparepvec (T-VEC; Imlygic) has been examined in combination with pembrolizumab. In a phase I study,2 the ORR with the combination was 62% for patients with melanoma. The phase III MASTERKEY-265/KEYNOTE-034 trial is currently examining this approach (NCT02263508).
“The MASTERKEY-265 frontline trial of T-VEC and pembrolizumab has now completed accrual,” Luke said. “We might see the first readout at ASCO next year.”
Another site-specific strategy for improving response to immunotherapy is looking at reactivation of the immune system following resistance. Intramural injections of the TLR9 agonists tilsotolimod (IMO-2125) was capable of triggering durable responses when used in combination with ipilimumab for patients with PD-1 resistant melanoma. The ORR with the combination was 39%.3 A phase III trial known as ILLUMINATE-301 is currently ongoing to further explore this combination (NCT03445533).
In addition to tilsotolimod, a number of other TLR9 agonists are under investigation, namely SD-101 and CMP-001, Luke noted. In addition to this, the STING pathway is another potential target, with several ongoing phase I trials looking at pembrolizumab with a STING agent, including MK-1454 (NCT03010176) and MK-2118 (NCT03249792). At this point, the future for novel combinations remains bright.