EIKCS 2022: Biomarker Discovery Through TCR Sequencing

(UroToday.com) In the tenth session of the 2022 International Kidney Cancer Symposium (IKCS): Europe meeting focusing on biomarker discovery in renal cell carcinoma (RCC), Dr. Samra Turajlic presented on the value of T cell receptor (TCR)-related metrics for understanding and predicting the response of immune checkpoint inhibition in clear cell renal cell carcinoma.

She began by highlighting that, over the past 24 hours at the IKCS conference, one of the recurring themes has been the importance of identifying the right treatment for the right patient, at the right time. However, in kidney cancer, this is limited by the fact that we have no biomarkers to guide treatment decision making. Thus, while we have seen improved response rates to novel treatment approaches that have been adopted over the past 30 years, there is little ability to predict those patients who will have a therapeutic response, or resistance.

This is not to say that a variety of things have not been assessed. She highlighted data examining INDEL load, tumor mutational burden (TMB), and PBRM1 mutation, as well as gene expression signature profiling, which have to date not proven clinically useful. While each, or at least many, of these have an underlying biological rationale or link, they have not demonstrated clinical utility part of this is driven by an inability to externally validate following initial discovery. This problem has particularly affected gene expression profiling which has held significant hope but, unfortunately, to date, has failed cross-validation. 

She then moved to emphasize that T cells are the main mediators of anti-tumor immunity. The T cell receptor (TCR) is the primary way that these cells interact with antigen. TCR expansion happens as a result of primary following engagement with antigen. Further, as each TCR is unique to an antigen, the TCR profile may act as a “barcode”. This can allow for longitudinal assessment of changes that may be occurring in the tumor and its microenvironment due to selective pressure of treatment. Further, while we are unable to directly identify the specific antigen, assessment of the TCR can to some degree allow us to impute this. This is important in RCC as specific antigens of importance have yet to be discovered.

While this work is in its relative infancy in RCC, she emphasized that there is precedent with proven utility of TCR profiling in triple negative breast cancer and melanoma. However, this work is ongoing in RCC, based in part on her own assessment of the ADAPTeR study. The ADAPTeR trial sought to assess the role of nivolumab as a pre- and post-operative treatment in metastatic renal cell carcinoma. From a biomarker development perspective, the study employed multi-regional sampling including primary tumors and metastatic sites across a number of time points. This longitudinal sampling allows for adaptive and dynamic assessment of changes as a result of the selection pressures of therapy. Further multi-regional sampling from tumor and metastasis allows both assessment and accommodation of the effects of spatial heterogeneity. Importantly in assessing markers of response and resistance, both pre- and post-treatment biopsies are available. While the ADAPTeR study has utilized a variety of platforms to explore RCC biology, this presentation focuses on work utilizing TCRSeq.

Dr. Turajlic first noted that TCR repertoires are very heterogenous between biopsy sites with only one of their included patients having a consistent pattern in TCR profile between sites. Thus, when considering clinical applicability of these approaches, we need to be mindful of the potential for sampling bias. To mitigate some of these effects, the authors sought to identify TCR patterns based on peripheral blood samples. However, unfortunately, they found that the clonality of the blood samples did not reflect that of the primary tumor. Thus, TCR clonal expansion is limited to being intra-tumoral.

However, this intratumor TCR expansion (with increased clonality) prior to therapy was predictive of subsequent response to nivolumab. This implies that responders have evidence of pre-treatment expanded clones, likely due to endogenous priming. Even more interestingly, these TCR populations are maintained following treatment (in blue in figure below) in responders while, in non-responders, they are replaced by new clones.

Further, among treatment responders, there is evidence that these expanded-maintained TCR clones show evidence of more cluster structure with sequence similarity. Building on this, Dr. Turajlic’s group sought to assess whether they could extract the same single from multi-regional bulk RNAseq using data from the TRACERx Renal cohort. To do so, they used MiXCR to assess TCR clonal interference and repertoire clonality and diversity. Using MiXCR, they were able to recapitulate the TCRseq results from the ADAPTeR trial, though with somewhat less sensitivity using this bulk sequencing approach. As the TRACERx Renal cohort has more spatial diverse samples, there is substantial TCR clonal heterogeneity. To address this, they have been working on an approach utilizing representative sampling to mitigate spatial sampling bias.

Concluding, Dr. Turjalic emphasized that TCR sequences may act as molecular barcodes for T cell clones and, with longitudinal sampling, allow for monitoring of T cell behaviour. Further, they may be a potential biomarker for immunotherapy response and prognosis. This may be particularly relevant in the adjuvant setting in which, following nephrectomy, there is an ability for comprehensive tumor sampling and profiling. However, she emphasized that we will likely need to combine this information with other biomarkers to provide clinical actionability.