How Useful Are Immune Checkpoint Markers in Squamous Cell Carcinoma?

Most immune checkpoint markers for oral squamous cell carcinoma (OSCC) have been studied little, despite their anticipated prognostic capabilities, according to a recent systematic review published in Oral Diseases.

“Despite the importance of immune checkpoints in immunotherapy, the prognostic value of these molecules remains controversial in oral squamous cell carcinoma,” wrote the authors, led by Meri Sieviläinen, of the Department of Oral and Maxillofacial Diseases, Clinicum, at the University of Helsinki in Helsinki, Finland.

The 5-year survival rate for OSCC, about 50%, has stagnated in recent decades; the cancer kills more than 145,000 people annually. Thus, new treatments and therapies for this disease are direly needed.

Currently, two US Food and Drug Administration–approved immunotherapies that target immune checkpoints are available to treat OSCC: the programmed death 1 (PD-1)/programmed death ligand 1 (PD-L1) inhibitors nivolumab and pembrolizumab. Various other immune checkpoint inhibitors are in different phases of clinical trials.

Sieviläinen et al aimed to review all immune checkpoint molecules for OSCC prognostication in the literature between 1985 and 2017. They identified 284 studies from the Ovid Medline, Scopus, and Cochrane databases. A total of 25 studies that examined 12 immune checkpoints or modulators met inclusion criteria. Only 4 checkpoints-PD-L1, PD-1, IDO1, and B7-H3-were examined in more than one study, and 8 immune checkpoints were examined in only one patient cohort.

On the basis of at least one study, the investigators found that 7 immune checkpoints (ie, FKBP51, B7-H4, B7-H6, ALHD1, PD-L1, B7-H3, and IDO1) were correlated with reduced survival in OSCC patients. In addition, 5 molecules (ie, CTLA-4, TLT-2, VISTA, PD-L2, and PD-1) harbored no prognostic value.

Importantly, PD-L1 is the best studied immune checkpoint, yet its prognostic value in OSCC remains unclear. The reviewers found that PD-L1 is significantly linked to both better and worse outcomes. Its prognostic value depends on its cellular location, with cytoplasmic location correlated with improved overall survival (OS), and membranous location associated with worse OS.

Of all the molecules studied, B7-H3 had the most promising potential, despite being examined only twice in OSCC. Both studies supported the prognostic value of this molecule.

“Its exact mechanisms are still unknown but there is evidence for co-stimulatory and coinhibitory signalling for adaptive immune system activation under different tumour contexts,” wrote the authors.

During the past several years, more than 100 OSCC prognostic biomarkers have been suggested to be prognosticators, but none are in clinical use. One issue is that potential prognostic markers for OSCC lack validation. Furthermore, there is a dearth of prospective studies, and multicenter studies are low power.

“Further research on immune checkpoint of OSCC should consider well-designed studies (both retrospective and prospective) with appropriate multivariate analysis of large cohorts,” concluded the authors.

In an interview with Cancer Network, Jong Chul Park, MD, an assistant at Massachusetts General Hospital and an instructor at Harvard Medical School, expressed intrigue about the results of the Sieviläinen et al study.

“Introduction of immunotherapy with immune checkpoint inhibitors has changed the therapeutic landscape of advanced HNSCC. However, only a small subset of patients with HNSCC benefit from immunotherapy with overall response rate of 15%, which means that [a] majority of patients are exposed to treatment-related toxicity and high cost without benefit from current immunotherapy,” he said.

“Identification of reliable predictive marker[s] for optimal patient selection and development of novel therapeutic[s] to overcome low immunogenicity is urgently warranted. It will be interesting to see if these immune checkpoints and modulators expression signatures shown to be prognostic in this review also have predictive value to the checkpoint inhibitors and furthermore suggest novel combination immune therapeutic approach,” Park added.

Park specifically pinpointed the potential prognostic value of PD-L1. “Currently, PD-L1 expression and tumor mutational burden seem to have the strongest evidence as predictive markers,” he said. “In HNSCC [head and neck squamous cell carcinoma], recent data presented at ESMO [the European Society for Medical Oncology meeting] this month of pembrolizumab compared to standard chemotherapy as a first line treatment (KEYNOTE-048) in advanced HNSCC suggests PD-L1 expression as a biomarker to be used in patient selection.”