Expert Interview: Cancer Resistance to Immunotherapy


Every patient has the ability to respond to immune therapy but too often tumors fail to regress due to either primary or acquired resistance. Here, we discuss the mechanisms behind immune evasion and how understanding the barriers to effective treatment in each individual patient can create a roadmap to a more effective and personalized treatment strategy with the goal of prolonged survival.

We were fortunate to speak with Dr. Siwen Hu-Lieskovan, PhD, MD, in a discussion on cancer resistance to immunotherapy. She is the Director of Solid Tumor Immunotherapy at the Huntsman Cancer Institute, University of Utah with research focused on immunotherapy resistance and its underlying mechanisms. In participation with the National Cancer Institute and SWOG, Dr. Hu-Lieskovan oversees the sub-study portfolio for iMATCH, a precision medicine trial for biomarker stratification.


PerkinElmer: I’d like to begin by asking how immunotherapies have transformed the cancer treatment landscape?

Dr. Siwen Hu-Lieskovan: There are many different kinds of cancer therapies originating with chemotherapy, radiation, surgery, later targeted therapy. These traditional cancer treatments focus on the tumor. If 99.9% of cancer cells are killed by these treatments, but even one remains, the disease will resurface at a later time.

The difference of immunotherapy is that the treatment doesn’t focus on the cancer, the treatment focuses on each individual patient’s immune system by rallying the immune cells against the tumor with the hope of eradication. Checkpoint inhibitors were a breakthrough in immunotherapy by unleashing activated immune cells to elicit their cytotoxic effects. This class of immunotherapies is FDA approved in more than 10 indications as a result of prolonged overall survival of patients when compared to other treatment options. Immunotherapy can work, however, overcoming mechanisms of immunosuppression characteristic of tumor cells requires a highly specified approach based on the individual functioning of each patient’s immune system.


PerkinElmer: What factors would you say contribute to immunotherapy resistance?

Dr. Siwen Hu-Lieskovan: Each tumor type exhibits unique features, and unfortunately, immune resistance mechanisms are universal across all types of cancer. Immunotherapy resistance may be observed primarily or following treatment as tumor cells mutate and evolve in their ability to evade the immune system. There are many steps involved, with tumor immunogenicity as step one. The immune system can recognize the tumor, and after that, primes the immune system to attack. Success depends on how strong the immune system is, and how immune cells overcome the tumor microenvironment.

PerkinElmer: And how often does resistance therapy result in hyperprogression in tumor growth?

Dr. Siwen Hu-Lieskovan: That is an interesting question. There have been reports of hyperprogression, but even defining the term is difficult. If the patient does not respond to therapy, they have progressed. How fast is hyperprogression compared to progression? There isn’t a definitive number or percentage, so whether a patient progresses more rapidly or not, without a defining set of criteria, the concept itself is kind of a myth.

PerkinElmer: Regarding your research, how has it contributed to the understanding of resistance mechanisms?

Dr. Siwen Hu-Lieskovan: I am a translational researcher. We focused on what is going on with each individual patient as resistance to immunotherapy can be unique to each patient. Again, we’re treating the immune system, not the tumor. That’s why we collect a patient’s tumor and blood samples, to study their unique features and dynamic change before and after treatment, and correlate with response and risk of resistance and toxicity. Primary resistance is difficult to study, from a genomics standpoint because of the heterogeneity. Acquired resistance is easier to study on a single patient basis, if the patient initially responds to immunotherapy but later exhibit resistance, you can study the progressing tumor compared to the baseline and to assess the difference.

For example, while I was working with Dr. Antoni Ribas’ group at UCLA, we discovered loss of function mutations in beta 2-microglobulin (B2M) and Janus kinase (JAK) can confer resistance to anti-PD-1 therapy in melanoma patients, by studying paired tumors from patients who developed acquired resistance.


Homozygous loss of function mutations in JAK 1 and JAK 2 were observed to be associated with the interferon (IFN)-receptor pathway. IFN exhibits pro-apoptotic and anti-proliferative functions and is therefore an important immunogenicity activator. Understanding the impact loss-of-function mutations have on suppressing anti-tumor immunity allows for the improvement of better immunotherapy treatment approaches.

PerkinElmer: Are there diagnostic tests that can predict immunotherapy resistance?

Dr. Siwen Hu-Lieskovan: There are currently no diagnostics that are approved to predict resistance by the FDA. There are biomarkers approved to predict potential response. PD-L1 is the first FDA approved test to select patients for treatment with checkpoint inhibitors in a few tumor types but it’s a somewhat controversial predictive biomarker. Tumor mutational burden (TMB) is a marker measured by counting the number of mutations present in a tumor cell. A high TMB is associated with a higher possibility of responding to immunotherapy.

PD-1 checkpoint inhibitor is approved for patients with a TMB of more than 10, and for patients with tumors that has microsatellite instability, which tends to have higher TMB. This is a huge breakthrough, as these approvals are biomarker driven, not restricted by a particular tumor type.

As clinicians, we utilize genetic testing to better understand patients’ tumors for better management. A lot of these tests tell us what the tumor mutational burden is, what the PD-L1 expression is, is there inflammation present in the tumor, and information on what mutations may be present. Thisdata help to predict whether a patient may or may not respond to immunotherapy, and what could be the potential resistance mechanisms, but so far everything is speculation.


PerkinElmer: On the topic of inflammation — how does it act as an independent biomarker?

Dr. Siwen Hu-Lieskovan: PD-1 is an inflammatory marker because it is predictive of adaptive immune response. CD8 T cells are also informative as a predictor of adaptive immune response, in some cancers like melanoma it is more informative than PD-L1. Their presence alone makes it difficult to discern if they are acting simply as bystanders, or if they are actually doing something. There are more and more efforts to examine gene expression profiles of the tumor to determine the inflammation state of the tumor, and what could be defective in mediating an effective immune attack.


PerkinElmer: How can researchers and clinicians combat resistance to create better outcomes for patients?

Dr. Siwen Hu-Lieskovan: That is the goal. Antitumor immune response is a process with many steps involved. In what way can the immune system be primed to promote anti-tumor activity? How strong can the immune system be? How well can therapies overcome the tumor microenvironment and intrinsic tumor mechanisms? These questions contribute to the complexity in overcoming immunotherapy resistance. Using biomarker investigation, researchers and clinicians can determine what specific mechanisms may be contributing to immunotherapy resistance within the tumor microenvironment. Better outcomes are observed when relevant biomarker data is used to apply a combined therapeutic approach targeting the various mechanism that contribute to unregulated tumor progression.

PerkinElmer: And finally, how can combination therapy approaches be tested in a clinical setting?

Dr. Siwen Hu-Lieskovan: I am working with the National Cancer Institute to design a clinical trial platform (iMATCH) focused on using prospective biomarker stratification of patients into groups with similar resistance mechanisms for testing of combination immunotherapy. A difficult part of testing combination therapy is the diversity of resistance mechanisms across individual patients in the clinical trial setting, and we do not know how to select patients for combination therapy that only address one mechanism. In a standard approach, certain combinations may show promise in the Phase I space, but in larger Phase III studies where patients are randomized and blinded into treatment and placebo groups without biomarker selection, the trials would fail to meet the benchmark. Without biomarker selection, the success of combination therapeutic strategy is low, not because the therapies are not working, but we are not finding the patients who would more likely to benefit from the approach.

In the iMATCH platform we will use biomarkers not to predict response but to inform biology. TMB suggests tumor immunogenicity, inflammation score indicates the current tumor immune contexture, and we use this combination to categorize the patients into high/high, high/low, high/low, low/low groups. The response signal to combination immunotherapies might be different in these biological subgroups and it can help to find the patient population that is most suitable for a certain combination therapy development. The field is realizing that it is important to test combination therapy in a more individualized manner.


ImmunoMATCH or iMATCH is a precision trial being managed at SWOG with the National Cancer institute. Categories based on resistance mechanisms are organized into different groups through biomarker analysis of each participant’s tumor biology and intrinsic immune functioning. Using immune profile testing, researchers will group via resistance mechanisms for precision combination therapy treatment.