Mechanisms of Primary and Secondary Resistance to Immunotherapy

It is no understatement that immunotherapy has completely changed the landscape of cancer treatment. In areas such as melanoma, renal cell cancer and refractory Hodgkin’s lymphoma, immunotherapy has given hope to a previously hopeless cohort of patients. It has also contributed to smaller - but no less meaningful - improvements in the treatment of lung, breast, gastric and cholangiocarcinomas.

However, the development and widespread use of immune checkpoint inhibitors (ICIs) has not been without its disappointments, as its impact on the treatment of pancreatic, proficient mismatch-repair colorectal cancer and prostate cancer has been negligible. In addition, primary or secondary resistance to ICIs can be devastating; in many cases, post-ICI treatment options remain very limited, with enrolment in clinical trials frequently the only option. As a result, it is critical - now more than ever - to improve our understanding of resistance mechanisms to ICIs. 

Primary resistance refers to a cancer that demonstrates no response to ICIs, suggesting the presence of pre-existing, intrinsic resistance mechanisms. Cancers that demonstrate hyper-progression fall into this category. In cancer types with a low response rate to ICIs, the incidence of primary resistance can be as high as 60%.

Secondary resistance refers to cancer that initially responds to immunotherapy, but subsequently develops acquired resistance, leading to growth and failure of treatment.

This summary uses material from two review articles: the first published by Drs Tiffany Set, Danny Sam, and Minggui Pan in 2019 and another published by Bai, Chen, and Li in 2020.

Image courtesy of Journal of Medial Sciences (2019)

Host Factors

Multiple studies in both mice and humans have demonstrated a correlation between healthy gut flora and response to ICIs. Mice with melanoma were noted to exhibit a muted response to anti-CTLA4 blockade if they were treated with antibiotics or fed with sterilised, germ-free diets. This response was restored when they were provided with Bacteroides fragilis. In humans with melanoma, the presence and relative abundance of the Ruminococcaceae family of normal flora was associated with a better response rate and longer progression-free survival (PFS). Gut microbiota is thought to have a regulatory role in the host’s systemic immune state. It has also raised questions regarding dietary modification for patients receiving ICIs, specifically the benefits of vegetarian diets or the use of probiotics, though no consensus is yet available.

Similarly, antibiotic use could be associated with compromised efficacy of ICIs through alteration of gut flora. Studies from Britain and France have demonstrated that antibiotic use before or during treatment with ICIs is associated with shorter PFS and overall survival (OS). However, it goes without saying that antibiotics are usually required to treat acute infections and complications, but these data support more judicious antibiotic use in patients receiving treatment with ICIs.

Steroid use remains controversial despite its impact on ICI therapy making both mechanistic and practical sense. A retrospective review of patients with metastatic non-small cell lung cancer (NSCLC) from the Sloan Kettering Cancer Center and Gustave Roussey Cancer Centre demonstrated that long-term administration of 10mg prednisolone daily (or equivalent) was associated with worse outcomes. However, a subsequent meta-analysis by Garant et al. found no difference in patient outcomes. Practically speaking, limiting the use of steroids in patients receiving ICIs should be encouraged, if only to avoid the significant issues caused by long-term steroid use.

Performance status (PS) and comorbidities are also likely to impact the efficacy of ICIs, as both factors are linked to patients’ immune function. However, a meta-analysis by Bersanelli et al. from 2018 demonstrated no association between poor PS and OS. The authors conclude that interactions between ICI efficacy and patient PS are more complex than those with chemotherapy.

Finally, several host biomarkers have been purported to offer windows into a tumour’s behaviour and speed of growth. Data published by Set et al. suggest that in a cohort of melanoma patients, elevated baseline neutrophils (>5500/microL), platelets (>304,000/microL), lactate dehydrogenase (LDH) and decreased lymphocyte count (<1120/microL) were associated with more rapid disease progression and poorer outcomes. Meanwhile, a baseline elevated lymphocyte count (>1716/microL) and a reduction in LDH following treatment were associated with better outcomes. This supports existing evidence that elevated neutrophil-to-lymphocyte ratio (NLR) is associated with significantly worse outcomes, while subsequent reduction of NLR with treatment is associated with a better patient outcomes.

Tumour Factors

Image courtesy of Frontiers of Oncology (2020)

Tumour factors in the development of resistance to ICIs remain highly complex. The tumour microenvironment frequently involves the expression of anergic and immunosuppressive proteins such as PD-L1, indoleamine-2,3-dioxygenase (IDO), and immunosuppressive cytokines such as transforming growth factor (TGF)-beta. In addition, tumour genetic factors such as aneuploidy (the expression of an abnormal number of chromosomes) are thought to render a tumour cell immunogenically “cold”, meaning that fewer markers that allow for immune cell infiltration are expressed.

The importance of tumour mutational burden (TMB), tumour-infiltrating T-lymphocytes (TILs), and other immune cells such as macrophages and dendritic cells are becoming more apparent as potent upregulators of immune response. Cancers with high TMB, deficiencies of mismatch repair proteins (dMMR) or a high degree of microsatellite instability (MSI-H) have demonstrated susceptibility to ICIs across multiple cancer subtypes, most notably in colorectal and endometrial cancers. These are associated with increased neoantigen expression and T-cell infiltration. However, as is always the case, cancers have multiple escape strategies: some cancer types can shed antigens to facilitate immune escape; others can undergo "antigenic drift”, similar to the annual influenza virus, to escape an established adaptive immune response. There is some association between the degree of PD-L1 expression, neoantigen expression and TILs infiltration, though the significance of this on tumour response to ICIs remains uncertain.

Thousands of factors are thought to contribute to an immunosuppressive microenvironment, both within and outside the tumour. Secretion of PD-L1 into peripheral blood, upregulation of the IL-6/JAK pathway, development of a hypoxic acidic tumour microenvironment (TME) through rampant metabolism, and overproduction of angiogenic factors such as vascular endothelial growth factor (VEGF) all serve to upregulate regulatory T cells (Treg), facilitate T-cell exhaustion, and downregulate the expression of CD8+ and natural killer (NK) cells.

Mechanisms of Primary Resistance

The most critical factors associated with primary resistance to ICIs are tumour biology and microenvironment. For example, resistance to the interferon (IFN)-gamma pathway has been associated with primary resistance, as illustrated below:

Image courtesy of Frontiers of Oncology (2020)

IFN-gamma production is associated with recruitment, activation, and augmentation of immune cells to the TME. It is also involved in the upregulation of genes associated with the production of MHC class 1 proteins, thus increasing the cancer cell’s antigenicity, in effect making it appear “more foreign” to the host’s immune system.

Host factors associated with primary resistance include previous organ or bone marrow transplantation, pregnancy, and chronic viral, bacterial, or fungal infections.

Mechanisms of Secondary Resistance

Mechanisms of secondary resistance are primarily an evolutionary consequence of tumour cells, whether by selection or adaptation in the face of immune-mediated destruction. Increased PD-L1 expression can overcome PD-L1 blockade, and the production of increased volumes of immunosuppressive cytokines or T-cell inhibitory receptors are all associated with the development of secondary resistance.

Tumours can also acquire mutations in the IFN-gamma signalling pathway as described above. Specific mutations include deletion of JAK1 and JAK2 genes, truncating mutations of the antigen-presenting protein Beta-2 microglobulin and the loss of PTEN. These mutations were all associated with decreased infiltration by immune cells and subsequent development of secondary resistance.

The Bottom Line:

It is often said that maximising the efficacy of an older anti-cancer treatment is more efficient than creating a new treatment modality from whole cloth. For example, the development of antibody-drug conjugates (ADCs) could be seen as a way of improving cytotoxic chemotherapy agents, which could be decades old.

Improving our knowledge of immune resistance mechanisms is an incredibly exciting area of research and has the potential to greatly increase the longevity of existing treatments and open doors to exciting new targets for anti-cancer therapy. The breadth and depth of research in this area are truly staggering, and this review cannot hope to summarise it adequately. However, it is exciting to imagine a world where personalised anti-cancer treatment is a process that occurs throughout a patient’s cancer journey, and “progressive disease” is no longer the gut-punch it is at present.

We thoroughly recommend reading the articles below for a more in-depth review of this fascinating topic.

Sources:

Seto T, Sam D, Pan M. Mechanisms of Primary and Secondary Resistance to Immune Checkpoint Inhibitors in Cancer. Med Sci. 2019, 7, 14; doi:10.3390/medsci7020014

Bai R, Chen N, Li L. Mechanisms of Cancer Resistance to Immunotherapy. Front. Oncol., 06 August 2020. Sec. Cancer Immunity and Immunotherapy; doi:10.3389/fonc.2020.01290

Mojic M, Takeda K, Hayakawa Y. The Dark Side of IFN-γ: Its Role in Promoting Cancer Immunoevasion. Int J Mol Sci. 2018 Jan; 19(1): 89. doi:10.3390/ijms19010089