Quest for Immune Checkpoint Inhibitor Biomarkers–a Long but Exciting Road

Targeted Therapies in Oncology, July 1 2021, Volume 10, Issue 9
Pages: 97

T-cell–targeted immunomodulators burst onto the scene a decade ago, and investigators turned their focus to immune checkpoints such as CTLA-4, PD-1, and PD-L1 to develop new cancer-fighting strategies.

When historians look back at this era of cancer care, it is quite possible that the development of immune checkpoint inhibitors (ICIs) will be seen as the beginning of a crucial new treatment paradigm. T-cell–targeted immunomodulators burst onto the scene a decade ago, and investigators turned their focus to immune checkpoints such as CTLA-4, PD-1, and PD-L1 to develop new cancer-fighting strategies.

As Caroline Robert, MD, PhD, head of the Dermatology Unit at Gustave Roussy institute and codirector of the Melanoma Research Unit at INSERM 981 Paris-Sud University near Paris, France, noted in a 2020 article in Nature Communications, the introduction of the CTLA-4–targeted ipilimumab (Yervoy) in 2011 sparked a race to develop more ICIs.1 By the end of 2020, more than 3000 active clinical trials were evaluating T-cell modulators, comprising two-thirds of all oncology trials.

Another 6 ICIs, all targeting the PD-1/PD-L1 signaling pathway, had been approved by the FDA as of December 2020. The approved indications range across 19 cancer types and 2 tissue-agnostic signatures.2

Scott M. Lippman, MD, professor of medicine, senior associate dean and associate vice chancellor for cancer research and care, and director of Moores Cancer Center at the University of California (UC) San Diego Health in California, said ICIs have been transformative because in many cases, patients who previously had very grim prospects were suddenly cured.

“There are 2 major things that are different from any other kind of systemic therapy we’ve had,” Lippman, who also holds the Chugai Pharmaceutical Chair in Cancer at UC San Diego Health, told Targeted Therapies in Oncology. “One is the broad spectrum of activity—lung, head and neck, all of them, and unselected patients. And 2 is the tail on the curve.”

However, if the tail on the curve—the limited number of patients whose survival is substantially improved by ICI therapy—is an indication of the tremendous possibilities with ICIs, it is also a sign of the current limitations of these therapies.

Jiuwei Cui, MD, of the First Hospital of Jilin University in Changchun, China, addressed the problem in a 2020 review article.3

“Although the clinical development of immune checkpoint inhibitor therapy has ushered in a new era of antitumor therapy, with sustained responses and significant survival advantages observed in multiple tumors, most patients do not benefit,” Cui and colleagues wrote.

Investigators know ICIs can work wonders; they just do not know which patients are most likely to benefit from these therapies.

Lippman specializes in head and neck cancer, and he has seen anti–PD-1 checkpoint inhibitors lead to dramatic results. However, they work in only approximately 20% of patients. Measuring PD-L1 protein expression on tumor cells can help identify patients that are likely to do well, but according to Lippman it is essentially a crude tool, and one that applies only to treatment with PD-1/PD-L1 inhibitors.

“PD-L1 is out there. It’s not great, as it applies to the PD-1 axis only,” he said. “So the question is, can we come up with better predictors of response and survival? And it’s these predictive genomics that will take [ICIs] into the era of precision therapy.”

First Wave of Biomarkers As Cui and colleagues noted, multiplex immunohistochemistry, high-throughput sequencing, and microarray technology have aided the ongoing development of a variety of biomarker strategies. The identification of single biomarkers or combinations of predictive markers helps enable oncologists to choose the most effective treatments for patients.3 The search for biomarkers has been wide ranging, though this research is very much in the early stages.

In a 2019 review article, Michael J. Duffy, PhD, and John Crown, PhD, of St Vincent’s University Hospital in Dublin, Ireland, outlined some of the key predictive biomarkers associated with response to ICIs. The authors also described several emerging biomarkers that are still undergoing validation but may provide valuable assistance with clinical decision-making in the future (TABLE).4 In addition to PD-L1 expression, microsatellite instability/defective mismatch repair (MSI/dMMR) and tumor mutational burden (TMB) were among the first ICI biomarkers to show promise.

A high level of MSI/dMMR is thought to be a pan-cancer biomarker that can be predictive of response to a specific ICI, such as pembrolizumab (Keytruda).

“MSI/dMMR is approved for clinical use irrespective of the tumor type, whereas PD-L1 is approved only for use in certain cancer types,” Duffy and Crown wrote.

At the time they were writing, in 2019, TMB had not yet been approved for clinical use as a biomarker. That changed in 2020, when the FDA approved pembrolizumab to treat metastatic or unresectable solid tumors with high TMB.

In a March 2021 review article, the FDA’s Julianne D. Twomey, PhD, and Baolin Zhang, PhD, explained that TMB, a broad assessment of the rate of mutations present in a tumor, appeared to be associated with ICI efficacy.2

“The higher mutational burden within a tumor is expected to correspond to a higher level of immunogenic neopeptides that would drive T-cell–mediated antitumor immunity,”they wrote.

Unfortunately, the logic of higher TMB corresponding to higher immunogenicity does not always hold. Naiyer A. Rizvi, MD, then of Columbia University, and colleagues discussed this problem in the American Society of Clinical Oncology Educational Book.6

“TMB is not a useful marker in all cancer types,” they wrote. “In some histologies, substantial responses to ICIs are observed despite a low TMB.”

For instance, they noted, in renal cell carcinoma,there does not appear to be a clear correlation between TMB and response to ICIs.

Additional Potential BiomarkersIn addition to PD-L1, MSI/dMMR, and TMB, a number of other potential biomarkers have been proposed. Serine/threonine kinase 11 (STK11) inactivation has been linked with a worse prognosis and resistance to ICIs in patients with KRAS-mutant lung cancers, Rizvi and colleagues reported.6

Additional Potential Biomarkers

In addition to PD-L1, MSI/dMMR, and TMB, a number of other potential biomarkers have been proposed. Serine/threonine kinase 11 (STK11) inactivation has been linked with a worse prognosis and resistance to ICIs in patients with KRAS-mutant lung cancers, Rizvi and colleagues reported.6

Mutations in Kelch-like ECH-associated protein 1 (KEAP1), which commonly co-occur with STK11 mutations, appear to correlate with poor clinical outcomes, according to results of a pan-cancer analysis published in 2020 in Annals of Translational Medicine.7 In a cohort of 1661 patients who received ICI therapy, the 99 patients with KEAP1 mutations had an over-all survival of just 10 months vs 20 months in the KEAP1 wild-type group (P = .0029).

Yet it is not clear whether these mutations can predict ICI-specific outcomes. For instance, results of a 2020 analysis of more than 2000 patients in a real-world non–small cell lung cancer (NSCLC) cohort suggested that the presence of STK11 and/or KEAP1 mutations conferred a worse prognosis regardless of the type of frontline therapy the patient received.8

“Mutations in STK11 or KEAP1 were associated with poor outcomes across multiple therapeutic classes and were not specifically associated with poor outcomes in [immune checkpoint blockade] cohorts,” wrote corresponding author Alice M. Walsh, PhD, of Bristol Myers Squibb.

Data from the KEYNOTE-042 study (NCT02220894) showed similar efficacy of pembrolizumab in patients with advanced PD-L1–positive NSCLC regardless of STK11 or KEAP1 mutational status.

The tumor suppressor PTEN has also been identified as a potential biomarker. Rizvi and colleagues noted that genetic aberrations in PTEN appear to be most common in endometrial cancer (66% of cases) but have also been found in 10% to 15% of cases in several other cancer types. A lack of PTEN expression is believed to impair T-cell–mediated antitumor activity, and research is ongoing to find out whether, and how, loss of PTEN protein affects ICI response.6

Upregulated WNT/β-catenin signaling and β2-microglobulin (B2M) mutation are 2 other possible biomarkers of ICI resistance, though research is in its early stages. As a part of the WNT signaling pathway, β-catenin is involved in cell differentiation and migration, whereas B2M plays a critical role in antigen presentation to cytotoxic T cells.

“Given its multitude of functions, dysregulation of the [WNT/β-catenin] pathway is frequently involved in tumorigenesis and cancer metastasis,” Rizvi and colleagues wrote. Similarly, they noted that B2M loss is a mechanism of immune evasion, and B2M mutations have been linked with resistance to anti–PD-1 therapy in several cancer types.

Zeroing in on 9p Deletion

What is notable about the list of biomarkers under investigation is not so much that it is long but, rather, that it has the potential to become much longer.

Lippman’s excitement about ICI biomarkers is heightened in part because of a discovery he and his colleagues made in a study whose results were published earlier this year.10 While researching links between genetic abnormalities and invasive head and neck cancers, they zeroed in on gene copy number imbalances that seemed to be involved in tumor development. However, they soon realized that the abnormalities they had discovered were also related to ICI susceptibility.

Lippman and colleagues found that loss of chromosome 9p and JAK2–PD-L1 codeletion were linked with ICI resistance.10 Lippman said the identification of a potential biomarker came as a welcome surprise.

“This 9p story was not our hypothesis going into [this study]; it just fell out of it,” he said.

Lippman said that in earlier research, his team had found evidence suggesting chromosome 3p loss and chromosome 9p loss played important roles in the development of human papillomavirus–negative head and neck cancers. For many years, 3p received most of the attention. That is, until Lippman’s new data came out. When the new data suggested 9p was the more important of the 2 types of deletion, Lippman and col-leagues knew the clinical implications, so they verified their findings using real-world post-ICI patient survival data from Caris Life Sciences. The association between chromosome 9p loss and ICI resistance held firm.

“The 9p observation was…a year and a half before we submitted the paper,” Lippman said. “It was so striking that we just wanted to make sure that we were 100% confident in that finding because it changed everything.”

Translating to the Clinic

Though the FDA has taken initial steps to pair approval of tools such as biomarker tests with therapeutic approvals, the type of routine testing and therapy selection envisioned by Lippman and others cannot yet happen on a wide scale due to the absence of 2 key factors: standardization and approved diagnostic platforms.

In their analysis of biomarkers, Duffy and Crown noted that although TMB, PD-L1, and MSI/dMMR have become important metrics, the utility of those biomarkers remains limited by the absence of clearly standardized assays. They added that uncertainties about appropriate cutoff values for biomarker positivity remain largely unresolved.4

“Despite being widely investigated, assays for MSI/dMMR, PD-L1, and TMB lack standardization and are still evolving,” Duffy and Crown wrote. “An urgent focus of future research should be the optimization and standardization of methods for determining these biomarkers.”

Lippman commented on the need for several additional measures, including FDA approval of more diagnostic assays to guide the use of ICI treatments. Still, he thinks our current path could lead to more advanced precision immunotherapy and the kind of world where next-generation sequencing informs therapeutic choices with very high rates of success. Such a future is in sight, he said, but there is still a long way to go.

“I really do think we’re going to eliminate cancer as we know it today,” Lippman said. “It’s not suddenly going to go away, but it’s not going to be the scary thing that it was before.”


1. Robert C. A decade of immune-checkpoint inhibitors in cancer therapy. Nat Commun. 2020;11(1):3801. doi:10.1038/s41467-020-17670-y

2. Twomey JD, Zhang B. Cancer immunotherapy update: FDA-approved checkpoint inhibitors and companion diagnostics. AAPS J. 2021;23(2):39. doi:10.1208/s12248-021-00574-0

3. Bai R, Lv Z, Xu D, Cui J. Predictive biomarkers for cancer immunotherapy with immune checkpoint inhibitors. Biomark Res. 2020;8:34. doi:10.1186/s40364-020-00209-0

4. Duff y MJ, Crown J. Biomarkers for predicting response to immunother-apy with immune checkpoint inhibitors in cancer patients. Clin Chem. 2019;65(10):1228-1238. doi:10.1373/clinchem.2019.303644

5. FDA approves pembrolizumab for adults and children with TMB-H solid tumors. FDA. Updated June 17, 2020. Accessed June 4, 2021.

6. Lagos GG, Izar B, Rizvi NA. Beyond tumor PD-L1: emerging genomic bio-markers for checkpoint inhibitor immunotherapy. Am Soc Clin Oncol Educ Book. 2020;40:1-11. doi:10.1200/EDBK_289967

7. Chen X, Su C, Ren S, Zhou C, Jiang T. Pan-cancer analysis of KEAP1mutations as biomarkers for immunotherapy outcomes. Ann Transl Med. 2020;8(4):141. doi:10.21037/atm.2019.11.52

8. Papillon-Cavanagh S, Doshi P, Dobrin R, Szustakowski J, Walsh AM. STK11 and KEAP1 mutations as prognostic biomarkers in an observational real-world lung adenocarcinoma cohort. ESMO Open. 2020;5(2):e000706. doi:10.1136/esmoopen-2020-000706

9. Cho BC, Lopes G, Kowalski DM, et al. Relationship between STK11 and KEAP1 mutational status and effi cacy in KEYNOTE-042: pembrolizumab monotherapy versus platinum-based chemotherapy as first-line therapy for PD-L1-positive advanced NSCLC. Cancer Res. 2020;80(suppl 16):CT084. doi:10.1158/1538-7445.AM2020-CT084

10. William Jr WN, Zhao X, Bianchi JJ, et al. Immune evasion in HPV–head and neck precancer–cancer transition is driven by an aneuploid switch involving chromosome 9p loss. Proc Natl Acad Sci U S A. 2021;118(19):e2022655118. doi:10.1073/pnas.2022655118