Cancer Immunotherapy Biomarkers

An article by John Haanen released in occasion of ESMO Immuno-Oncology Congress 2017, 7-10 December, Geneva, Switzerland.

Cancer Immunotherapy Biomarkers

Immunotherapy has revolutionised the treatment of many advanced cancers including melanoma and non-small cell lung cancer (NSCLC). In particular, treatment with immune checkpoint inhibitors—(ICPis)—monoclonal antibodies that block cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) and the programmed death-1 (PD-1) receptor (PD-1) and its ligand, PD-L1—offer durable tumour regressions in a substantial proportion of patients across a wide range of advanced cancers (1). Response rates are variable, with about half of patients with advanced melanoma and a larger proportion of NSCLC patients, for example, not responding to ICPis. Thus, identifying biomarkers to predict benefit from ICPis is an area of intense interest.

The complexity and heterogeneity of a tumour cell-immune system interaction ultimately determines responsiveness (2). The cancer immunogram that was developed to describe the cancer-immune system interaction is a framework of seven parameters that are known to affect the anticancer immune response and that can be interrogated in individual cancer patients (2). These parameters are: tumour mutational burden; the general immune status of the patient; presence of T cell immune infiltrates; tumour PD-L1 expression; sensitivity of tumour cells to T-cell killing (including MHC expression, functional IFN-g receptor pathway); a myeloid cell-mediated inflammation (high C-reactive protein (CRP) and IL-6 levels); and high serum lactate dehyrogenase (LDH) (reflecting both tumour burden and anaerobic glycolysis). Based on currently available data from clinical and translational research studies, highly predictive biomarkers of response will be multifaceted and probably differ between tumour types. We will describe and discuss the presently available data on several of the biomarkers involved in the success of immunotherapy of cancer.

PD-L1 expression

Binding of PD-L1, the main ligand of PD-1, inhibits the function of activated T-cells (3). A range of solid and haematological malignancies express PD-L1, co-opting the PD-1/PD-L1 immune response checkpoint to evade immune attack (1). PD-L1 expression can be caused by an oncogenic event, such as gene amplification, which occurs in among others Hodgkin’s lymphoma. More commonly, it is the result of an adaptive immune resistance mechanism—recognition of tumour cells by activated antigen-specific PD-1-expressing T-cells, which leads to IFNg release and in turn, through IFN-g receptor signalling in tumour cell PD-L1 expression that via PD-1-PD-L1 interaction shuts down this activated T cell response (3).

For many tumour types, expression of PD-L1 is correlated with response to PD-1/PD-L1 inhibition, but the required expression level varies from >1% to 50% or more (3). In melanoma, NSCLC, and bladder cancer, PD-L1 immunohistochemistry (IHC) has identified patients with a higher likelihood of treatment response to anti-PD-1/PD-L1. Importantly, in all studies patients with very low or no PD-L1 expression on tumour cells still derive some benefit from anti-PD-1/PD-L1 treatment, arguing for the poor predictive value of PD-L1 expression as a biomarker for outcome. Pembrolizumab was the first checkpoint inhibitor to be approved with the requirement for a diagnostic test to measure tumour PD-L1 expression (PD-L1 IHC 22C3), as a single agent for NSCLC treatment. Patients with a high tumour PD-L1 expression, as defined by a proportional score of ≥50 % (PS ≥50 %), demonstrated improved objective response rates and survival in NSCLC patients compared to standard of care chemotherapy (4). However, in another frontline randomised, controlled phase III study with nivolumab in NSCLC, expression of PD-L1 was not associated with outcome (CheckMate 026), illustrating the variability in the predictive value of this biomarker (5). The finding that tumours that are IHC-negative for PD-L1 can still respond to PD-1/PD-L1 blockade undermines its use as a treatment selection assay (3). The response from apparently PD-L1 negative tumours may reflect issues with the IHC testing and tumour biology (limitations in tumour sampling, such as focal expression, dynamic PD-L1 expression over time and by anatomical site and different IHC detection methods and antibodies). Currently, four commercial IHC tests have been approved by the US Food and Drug Administration (FDA) as either companion or complementary diagnostics for anti-PD-1/PD-L1 therapies: PD-L1 IHC 22C3, PD-L1 IHC 28-8, PD-L1 IHC SP142, and PD-L1 IHC SP263. Recently, comparison of the different IHC tests and cell scoring methods for PD-L1 expression showed that the percentage of PD-L1-positive tumour cells was comparable when the 22C3, 28-8, and SP263 assays were used, whereas the SP142 assay exhibited fewer stained tumour cells (6).

Tumours with high mutational burden—microsatellite instability (MSI)-high or DNA mismatch repair deficient (dMMR)

Tumour response to anti-PD-1 blockade occurs more frequently in tumours with a high mutational burden, although a specific cut-off in the number of mutations required for an effective immune response could not be defined, making mutational burden not a very precise biomarker. The hypothesis is that a high mutational burden increases the foreignness of cancers. A high number of mutations will increase the probability of neoantigen generation—mutated epitopes generated from mutated expressed genes—that when bound to MHC molecules can be seen by T-cells (7). Generation of neoantigens is like a lottery: the more mutations, the higher the chance for a neoantigen, but one ‘good’ neoantigen can be enough for an effective anti-tumour response.

Correlation between outcome and mutational burden has been shown for melanoma, NSCLC, urothelial cancer and for tumours with defects in DNA repair mechanisms. Especially, patients with either germline or somatic mutations in mismatch repair genes, have a very high number of non-synonymous mutations, reflected by unstable microsatellite regions in the genome. Irrespective of the origin of the cancer, patients with mismatch repair deficient tumours (dMMR) have a high overall response rate to PD-1/PD-L1 blockade. This is now reflected in the US FDA approval of pembrolizumab for any advanced tumour with MSI-high or dMMR, independent of tumour type or site. The indication is for adult and paediatric patients with unresectable or metastatic solid tumours that have progressed on alternative drugs, based on a positive IHC or PCR test for MSI-high or dMMR. Five single-armed studies involving 149 patients with 15 types of solid tumours reported a 40% complete or partial response (1).

Other outcome biomarkers for immune checkpoint inhibitors

T-cell receptor repertoire

The T-cell repertoire needs to be broad enough to induce or mobilise an endogenous immune response; a limited T-cell receptor (TCR) repertoire would decrease the likelihood of generating a cancer-specific T cell response. By DNA sequencing of the TCR Vβ genes, the diversity of the TCR repertoire can be assessed. This has been studied both in melanoma patients treated with ipilimumab (8) and in bladder cancer patients treated with atezolizumab (9). The presence of a restricted or uneven TCR repertoire measured in the peripheral blood had a negative impact on survival upon treatment with these ICPis.  At the same time, a more clonal T-cell response at the tumour site was correlated with a better outcome to these antibodies. The hypothesis is that clonal expansion of tumour resident T-cells reflects an ongoing tumour-specific immune response.

Tumour microenvironment and the tumour gene expression profile

The nature of the tumour microenvironment also plays an important role in ICPi response (10). Tumours can be categorised as “inflamed” or “hot” tumours, showing immune cell infiltrates within the central part of the tumour or in the invasive tumour margins, tumour PD-L1 expression, and an IFNg or T-cell gene signature with expression of genes that are correlated with an activated or cytolytic T-cell response. Other tumours are completely devoid of immune infiltrates, these are categorised as “immune desert” or “cold”. These tumours exhibit primary resistance to immunotherapy. The third category are tumours that are immune excluded, where immune cells reside in the tumour stroma but are incapable of entering the tumour; these tumours also may be resistant to the currently available immunotherapies. These categories do not express a T cell or IFNg gene signature. For these tumours to become responsive to immunotherapy, the barriers within the tumour microenvironment that hamper T cell infiltration need to be broken.

Serum biomarkers

Although the role of serum biomarkers as predictive outcome markers for ICPis therapy remains to be established (10), several serum markers such as lactate dehydrogenase (LDH), CRP, vascular endothelial growth factor (VEGF) and soluble CD25 are associated with response to anti-CTLA-4 and PD-1/PD-L1 blockade in advanced melanoma (11-15). These serum markers, especially LDH and CRP, are easy to use and can be implemented in choosing the best treatment option for patients with metastatic melanoma. Serum LDH >2xULN is highly predictive of unresponsiveness to CTLA-4 blockade, and to a lesser extend also to anti-PD-1 treatment. If feasible, these patients should be treated with combination immunotherapy with anti-CTLA-4 and anti-PD-1, to maximise the chance for response and durable outcome (16).

Baseline neutrophil-to-lymphocyte ratios and relative eosinophil count are associated with response to ICPi, although none of these markers have been demonstrated to reliably identify a patient who will not benefit from treatment as yet (15).

Opportunities and challenges for biomarker development

While the complexity of the human immune system poses a challenge to the development of biomarkers, the availability of powerful technologies and high throughput approaches that include mass cytometry, whole exome sequencing, and gene expression profiling offer important opportunities for biomarker development (17). Challenges remain; larger clinical studies are required to determine the real predictive value of these promising parameters and further understanding of the interactive and dynamic nature of the immune response is required to define the best biomarkers of response to PD-1/PD-L1 checkpoint inhibitors. It is likely that more complex predictive biomarker systems based on tumour cell-and T-cell-specific markers will be developed and refine appropriate patient selection for PD-1/PD-L1 blockade.

References

  1. Topalian SL. Targeting immune checkpoints in cancer therapy. JAMA 2017; 318(17):1647-1648.
  2. Blank CU, Haanen JB, Ribas A, et al. CANCER IMMUNOLOGY. The "cancer immunogram". Science 2016; 352(6286):658-660.
  3. Patel SP, Kurzrock R. PD-L1 Expression as a predictive biomarker in cancer immunotherapy. Mol Cancer Ther 2015;14(4):847-856
  4. Reck M, Rodríguez-Abreu D, Robinson AG, et al. Pembrolizumab versus chemotherapy for PD-L1–positive non–small-cell lung cancer. New Engl J Med 2016; 375(19):1823–1833.
  5. Carbone DP, Reck M, Paz-Ares L, et al. First-line nivolumab in stage IV or recurrent non-small-cell lung cancer. N Engl J Med 2017; 376(25):2415-2426.
  6. Hirsch FR, McElhinny A, Stanforth D, et al. PD-L1 Immunohistochemistry Assays for Lung Cancer: Results from Phase 1 of the Blueprint PD-L1 IHC Assay Comparison Project. J Thorac Oncol 2017; 12(2):208-222.
  7. Hugo W, Zaretsky JM, Sun L, et al. Genomic and transcriptomic features of response to anti-pd-1 therapy in metastatic melanoma. Cell 2016; 165(1):35-44.
  8. van Rooij N, van Buuren MM, Philips D, et al. Tumor exome analysis reveals neoantigen-specific T-cell reactivity in an ipilimumab-responsive melanoma. J Clin Oncol 2013; 31:e439-442.
  9. Snyder A, Nathanson T, Funt SA, et al. Contribution of systemic and somatic factors to clinical response and resistance to PD-L1 blockade in urothelial cancer: An exploratory multi-omic analysis. PLoS Med 2017: 14(5): e1002309.
  10. Weber JS. Biomarkers for checkpoint inhibition. Am Soc Clin Oncol Educ Book. 2017; 37:205-209.
  11. Kelderman S, Heemskerk B, van Tinteren H, et al. Lactate dehydrogenase as a selection criterion for ipilimumab treatment in metastatic melanoma. Cancer Immunol Immunother 2014; 63(5):449-458.
  12. Yuan J, Zhou J, Dong Z, et al. Pretreatment serum VEGF is associated with clinical response and overall survival in advanced melanoma patients treated with ipilimumab. Cancer Immunol Res 2014; 2(2):127-132.
  13. Simeone E, Gentilcore G, Giannarelli D, et al. Immunological and biological changes during ipilimumab treatment and their potential correlation with clinical response and survival in patients with advanced melanoma. Cancer Immunol Immunother 2014; 63(7):675-683.
  14. Hannani D, Vetizou M, Enot D, et al. Anticancer immunotherapy by CTLA-4 blockade: obligatory contribution of IL-2 receptors and negative prognostic impact of soluble CD25. Cell Res 2015;25(2):208-224.
  15. Weide B, Martens A, Hassel JC, et al. Baseline biomarkers for outcome of melanoma patients treated with pembrolizumab. Clin Cancer Res 2016; 22(22):5487-5496.
  16. Wolchok JD, Chiarion-Sileni V, Gonzalez R, et al. Overall survival with combined nivolumab and ipilimumab in advanced melanoma. N Engl J Med 2017; 377(14):1345-1356.
  17. Gulley JL, Berzofsky JA, Butler MO, et al. Immunotherapy biomarkers 2016: overcoming the barriers. J Immunother Cancer 2017; 5(1):29.