ESOU 2022: Liquid Biomarkers in Advanced Kidney Cancer

( The 2022 EAU Section of Oncological Urology (ESOU) Annual Meeting included a session on locally advanced and metastatic renal cancer and a presentation by Dr. Lewis Au discussing liquid biomarkers in advanced kidney cancer. Dr. Au notes that the purpose of liquid biomarkers is to capture tumor genomic features through blood sampling. With regards to general nomenclature, Dr. Au notes that circulating tumor DNA (ctDNA) is not interchangeable with cell-free DNA (cfDNA), which is not interchangeable with circulating tumor cells (CTCs):


Lewis Au-0.jpg 

There are several advantages of serial liquid biopsies, including:

  • Cancer detection: screening or earlier diagnosis
  • Molecular profiling or prognostication
  • Detection of residual disease
  • Monitoring response
  • Monitoring clonal evolution

For translating tissue-based biomarkers of clear cell RCC to liquid biopsy assays, there are several important points to consider:

  1. ctDNA utility – the potential to change diagnostic and treatment approaches
    1. Detection of minimal residual disease in colorectal cancer
    2. FDA approved T790M EGFR mutant detection in NSCLC for resistance monitoring
    3. Prediction of relapse in breast cancer
    4. Depiction of cancer evolution in NSCLC
  2. Tissue-based genomic biomarkers (prognostic and predictive) in clear cell RCC:
    1. VHL (~90% mutated or methylated), PBRM1 (~40%), BAP1 (~15%), SETD2 (~15%)
    2. PBRM1 mutation is generally a favorable prognostic biomarker, but varied associations with response to VEGF or immunotherapies
    3. BAP1 and TP53 mutations are associated with poor prognosis
    4. SETD2 mutation is not consistently associated with prognosis or therapy outcomes

Dr. Au notes that there are several challenges with regards to clear cell RCC and liquid biomarkers. First, is intratumor heterogeneity. Work from Dr. Au’s group assessed determinants of anti-PD-1 response and resistance in clear cell RCC, published in Cancer Cell in 2021.1 Using the ADAPTeR phase II study of nivolumab in 15 treatment-naive patients (115 multiregion tumor samples) with metastatic clear cell RCC, genomic analyses show no correlation between tumor molecular features and response, whereas clear cell RCC-specific human endogenous retrovirus expression indirectly correlates with clinical response. Additionally, T cell receptor analysis revealed a significantly higher number of expanded T-cell receptor clones pre-treatment in responders suggesting pre-existing immunity. Maintenance of highly similar clusters of T-cell receptors post-treatment predict response, suggesting ongoing antigen engagement and survival of families of T cells likely recognizing the same antigens. These data suggest nivolumab drives both maintenance and replacement of previously expanded T cell clones, but only maintenance correlates with response.

The second challenge is low ctDNA levels in clear cell RCC. Compared to other malignancies, RCC ranks among the lowest for ctDNA levels:

Lewis Au-1.jpg 

The third challenge is that ctDNA levels track with clinical disease stage, as highlighted in the following figure showing increasing ctDNA levels with increasing stage:

Lewis Au-2.jpg 

The fourth challenge is deciphering biological noise in ctDNA analysis, given that >80% of cfDNA arises from hematopoietic cells, which accumulate somatic mutations during aging, leading to clonal expansion without dysplasia (clonal hematopoietic mutations of indeterminate potential – CHIP). Furthermore, biological noise confounds low-frequency variants, leading to false-positive plasma genotyping. Based on these four challenges, there is a need for high depth target capture sequencing, matched multiregion tumor mutation profiling and sequencing error suppression, and a longitudinal, phylogenetic approach for the evaluation of stage I-IV disease.

Dr. Au then discussed the first results of patients in Renal TRACERx, a prospective blood and multiregion tumor sampling study with plasma collection at 194 time points across stage I-IV disease:

Lewis Au-3.jpg 

There are several key aims within TRACERx Renal:

  1. Can ctDNA detect clonal and subclonal tumor mutations?
  2. Does the level of ctDNA correlate with tumor burden?
  3. Does the level of ctDNA track with response to therapy?
  4. Is detection of radiologically-occult disease feasible through ctDNA analysis?

As follows is the clinical cohort of the first 21 patients:

Lewis Au-4.jpg 

cfDNA is extracted from 4mL of plasma and next-generation sequencing profiling is undertaken by bespoke target capture enabling monitoring of multiple mutations and clonal evolution. Dr. Au notes that incorporation of unique molecular indexes reduces the risk of duplicate reads and errors, with sequencing to a target depth of 20,000x. Mutations inferred from a multi-region profiling of primary tumor tissues is used as a reference and the ground truth. Furthermore, the minimum variant allele frequency threshold is 0.10% for ctDNA.

The preliminary results of this study reported a mean extracted cfDNA concentration of 3.92 ng/uL (range 0.17 – 40.2 ng/uL), a median sequencing coverage of 11,360x (range 4,637-21,350x), and median duplication rate of 72% (range 45.9% - 93.9%). For stage I-II, no mutations were detected by ctDNA profiling, for stage III mutations were detectable in one patient (50% of relapse events), and for stage IV mutations were detectable in the majority (10/11) of cases:

Lewis Au-5.jpg 

The sensitivity of detection by cfDNA with regards to the overall number of mutations in the primary tumor was 72.7%, for clonal mutations were 90.9%, and for subclonal mutations was 36.4%. One PTEN mutation detected in a single patient at 4% VAF was not found in the primary tumor (97% specificity). Additionally, VAF dynamics mirror clinical response to therapy and track with disease progression, as well as VAF also potentially being able to predate radiological relapse:

Lewis Au-6.jpg 

Dr. Au concluded his presentation of liquid biomarkers in advanced kidney cancer with the following summary points:

  • Mutation tracking via ctDNA profiling has multiple applications for the clinical management of cancer
  • Next-generation sequencing can robustly detect tumor-specific mutations in ctDNA in clear cell RCC
  • Clonal phylogenies reconstructed from the multiregion sequencing data of the primary tumor allow tracking of clonal and subclonal mutations (phylogenetic tumor branches) in plasma
  • ctDNA variant allele frequencies can track therapy response
  • Given the treatment of advanced clear cell RCC in the adjuvant setting, minimal residual disease detection is a key research priority
  • Other liquid biomarkers include: (i) ctDNA (copy number profiling, fragmentation patterns, methylation profiling), (ii) PBMC/TCR sequencing, (iii) neutrophil-to-lymphocyte ratios, (iv) analysis of urine samples

Presented by: Lewis Au, Oncology Research Fellow, The Royal Marsden NHS Foundation Trust, Cancer Dynamics Laboratory, Francis Crick Institute, London, UK

Written by: Zachary Klaassen, MD, MSc – Urologic Oncologist, Assistant Professor of Urology, Georgia Cancer Center, Augusta University/Medical College of Georgia, @zklaassen_md on Twitter during the 2022 EAU Section of Oncological Urology (ESOU) Hybrid Annual Meeting, Madrid, Spain, Fri, Jan 21 – Sun, Jan 23, 2022.


  1. Au L, Hatipoglu E, de Massy MR, et al. Determinants of anti-PD-1 response and resistance in clear cell renal cell carcinoma. Cancer Cell. 2021 Nov 8;39(11):1497-1518.