ctDNA: An emerging neoadjuvant biomarker in resectable solid tumors

Guest editors Christopher Abbosh and Charles Swanton discuss circulating tumor DNA as a potential biomarker for neoadjuvant treatment response in solid tumors in a perspective article as part of a PLOS Medicine Special Issue: Early Detection and Minimal Residual Disease

Christopher Abbosh¹*, Charles Swanton^1,2

1. Cancer Research UK Lung Cancer Centre of Excellence, University College London Cancer Institute, London, UK
2. Cancer Evolution and Genome Instability Laboratory, The Francis Crick Institute, London, UK 

* c.abbosh@ucl.ac.uk

Introduction

Three studies presented within this special issue of PLOS Medicine focus on evaluation of circulating tumor DNA (ctDNA) as a response biomarker in early-stage solid tumours. Both Yaqi Wang and Pradeep Chauhan and their respective colleagues evaluate ctDNA as a tool capable of predicting complete pathological response (pCR) in locally advanced rectal cancer (LARC) and muscle invasive bladder cancer (MIBC), respectively [1,2]. Jeanne Tie and colleagues focus on ctDNA evaluation in high-risk metastatic colorectal cancer with liver metastases (CRLM), both during neoadjuvant therapy and following surgery and adjuvant therapy [3]. In this brief perspective we evaluate these advances within the wider context of recently published work.

ctDNA as a neoadjuvant response biomarker

Quantitation of ctDNA kinetics over time can act as a dynamic biomarker of tumour response to targeted therapies, immunotherapy and radiation therapy [4-6]. In a neoadjuvant setting ctDNA kinetics could guide escalation of neoadjuvant therapy in non-responders, or be used as a de-escalation tool to curtail the number of neoadjuvant cycles being administered or reduce the need for further therapy (including surgery).

In relation to the latter point, Wang and colleagues draw attention to the potential for deferral of surgery in patients exhibiting complete clinical response (cCR; i.e., no clinical, endoscopic, or radiographic evidence of disease) following neoadjuvant chemoradiation treatment for LARC [2]. This is termed a “watch and wait” strategy. In patients exhibiting cCR following neoadjuvant treatment, probability of LARC recurrence without further intervention is low [7]. However, cCR is an imperfect surrogate for pCR therefore the authors sought to determine whether ctDNA evaluation during neoadjuvant therapy could improve the accuracy of delineating treatment response. Through evaluation of a metric termed T234_clearance (describing absence of the highest mutant allele frequency baseline mutation at all 3 pre-operative timepoints) the team identified 20 of 89 patients who lacked evidence of T234_clearance following neoadjuvant treatment. Four of these patients exhibited cCR based on Magnetic Resonance Imaging (MRI) evaluation [2]. This observation highlighted some discordance between cCR and ctDNA clearance kinetics. Within the T234_clearance negative, cCR positive population, one of four patients suffered disease recurrence. This suggests that ctDNA clearance could refine cCR evaluation in LARC [2]. An exploratory analysis demonstrated that combining multiple ctDNA features alongside MRI response information in a prediction model improved discrimination of pCR from non-pCR, compared to ctDNA features or MRI response parameters alone [2].

The study from Wang and colleagues supports prior findings from Murahashi and colleagues who also evaluated post-treatment ctDNA kinetics in patients undergoing neoadjuvant treatment for LARC [8]. In Murahashi and colleagues’ analysis, post-treatment ctDNA levels decreased in 11 of 12 patients who experienced clinical response with neoadjuvant treatment (either pCR or 12 months relapse free if watch and wait strategy adopted); in contrast, 7 of 39 patients who did not experience clinical response displayed increase in ctDNA levels [8]. Combining decrease in ctDNA during treatment with endoscopic complete response evaluation improved neoadjuvant therapy response stratification in this study [8]. Overall, these data suggest that monitoring ctDNA following neoadjuvant treatment for LARC has potential as a complementary tool to improve accuracy of current cCR measures.

Like the application of ctDNA in LARC, Chauhan and colleagues asked whether evaluation of urinary ctDNA (utDNA) could differentiate pCR from non-pCR in patients being treated with neoadjuvant therapy for MIBC [1]. Development of an alternative to pCR in this setting could avoid surgical cystectomy and urinary diversion in excellent prognosis patients. The team identified that non-silent mutations (mutations that result in a change to an encoded amino acid sequence), but not silent mutations (mutations that do not change the amino acid sequence), were accurate determinants of pCR versus non-pCR and suggested that a field-effect within bladder urothelium could underlie this difference (a field-effect or field cancerisation describes tissue that has been pre-conditioned by carcinogen exposure, facilitating the process toward cancer formation [9]). Excluding silent mutations, a utDNA MRD threshold was optimised based on analyses of healthy participants and non-pCR patients. Application of this threshold to the cohort revealed a sensitivity of 81% and a specificity of 81% for non-pCR prediction. Based on these findings Chauhan and colleagues suggest utDNA could be used to complement emerging clinical predictors of cPR, such as MRI-based response criteria, and draw attention to the Alliance Study (clingov: NCT03609216) which will provide a framework for validation of the team’s observations. This work highlights the potential for urinary utDNA to be utilised as a biomarker in bladder cancer, building upon previous work using the same ctDNA platform (uCAPP-seq) by Dudley and colleagues that demonstrated utDNA as capable of identifying localised early-stage bladder cancer and tracking recurrent disease following local bladder cancer treatment [10].

Tie and colleagues explored ctDNA as a curative-therapy response biomarker in CRLM. Within a cohort of patients who underwent neoadjuvant chemotherapy, they noted a median 40.93-fold decrease in ctDNA levels during treatment with 13 of 18 evaluable patients exhibiting lack of ctDNA detection pre-cycle 4 of treatment: all 4 patients with pCR in the resection specimen experienced ctDNA clearance prior to cycle 3 or 4 of neoadjuvant treatment. However, ctDNA clearance during neoadjuvant chemotherapy had no impact on 5-year relapse free survival (RFS) when compared to lack of ctDNA clearance. In contrast, the team identified that ctDNA detection after surgery was a strong predictor of reduced RFS with patients who were ctDNA positive following curative therapy (surgery +/- adjuvant therapy) exhibiting a 5-year RFS rate of 0% versus 75.6% in ctDNA negative patients. These data highlight the importance of associating ctDNA clearance dynamics during neoadjuvant treatment with post-operative survival endpoints, since in this study ctDNA clearance with neoadjuvant chemotherapy did not translate into reduced risk of disease recurrence following surgery.

Supporting ctDNA as a neoadjuvant response biomarker in other tumor types, data in non-small-cell lung cancer (NSCLC) from the neoadjuvant CheckMate-816 study, a randomized, phase III study comparing neoadjuvant platinum chemotherapy with or without nivolumab in stage IB-IIIA NSCLC, highlighted that ctDNA clearance at day 1 cycle 3 post-combination chemotherapy and immune checkpoint inhibitor treatment associates with pCR [11]. Stage II-III early-stage breast cancer patients treated with either standard neoadjuvant chemotherapy or neoadjuvant chemotherapy plus MK-2206 (an AKT inhibitor) underwent longitudinal ctDNA-analyses using a tumour-informed assay in the I-SPY2 platform trial [12]. ctDNA clearance following cycle 1 of therapy associated with an increased likelihood of pCR (24 of 29 [83%] patients with non-pCR had residual ctDNA detected post cycle 1 of therapy versus 14 of 27 [52%] who cleared ctDNA post cycle 1 of therapy [12]). In this study the authors categorised patients by pCR status and ctDNA status after neoadjuvant therapy and identified that patients who were ctDNA negative but did not achieve pCR had a similar risk of metastatic recurrence compared with patients who did achieve pCR, suggesting that ctDNA clearance could divide non-pCR patients into high- and low-risk categories [12].

In conclusion, the findings presented in this special issue add to an emerging literature highlighting a need to explore the translational potential for ctDNA assessment as a response biomarker in the neoadjuvant setting. These data are particularly relevant in LARC and MIBC where treatment response biomarkers that are not reliant on pathological examination of resection specimens are required to guide non-operative management decisions. The data from Tie and colleagues suggest that the capability of neoadjuvant chemotherapy-induced ctDNA clearance to act as a surrogate of long-term survival benefit from curative intent therapy is limited in CLRM. It is conceivable that the utility of ctDNA as a neoadjuvant response biomarker may vary by therapeutic class and solid tumour type. To address this issue, it will be important for prospective interventional trials to incorporate ctDNA clearance kinetics as an endpoint to determine surrogacy of these measures for survival across solid-tumour types. Finally, to gain understanding of the relative merits and disadvantages of ctDNA-based response metrics versus conventional clinical measures of response (such as endoscopic and imaging-based evaluations), direct comparison of ctDNA clearance with these approaches is warranted.

References:

1.         Chauhan PS, Chen K, Babbra RK, Feng W, Pejovic N, Nallicheri A, et al. Urine tumor DNA detection of minimal residual disease in muscle-invasive bladder cancer treated with curative-intent radical cystectomy: A cohort study. PLoS Med. 2021;18(8): e1003732. https://doi.org/10.1371/journal.pmed.1003732

2.         Wang Y, Yang L, Bao H, Fan X, Xia F, Wan J, et al. Utility of ctDNA in predicting response to neoadjuvant chemoradiotherapy and prognosis assessment in locally advanced rectal cancer: A prospective cohort study. PLoS Med. 2021;18(8): e1003741. https://doi.org/10.1371/journal.pmed.1003741

3.         Tie J, Wang Y, Cohen J, Li L, Hong W, Christie M, et al. Circulating tumor DNA dynamics and recurrence risk in patients undergoing curative intent resection of colorectal cancer liver metastases: A prospective cohort study. PLoS Med. 2021;18(5): e1003620. https://doi.org/10.1371/journal.pmed.1003620

4.         Moding EJ, Liu Y, Nabet BY, Chabon JJ, Chaudhuri AA, Hui AB, et al. Circulating tumor DNA dynamics predict benefit from consolidation immunotherapy in locally advanced non-small-cell lung cancer. Nature Cancer. 2020;1(2):176-83. doi: 10.1038/s43018-019-0011-0.

5.         Raja R, Kuziora M, Brohawn PZ, Higgs BW, Gupta A, Dennis PA, et al. Early Reduction in ctDNA Predicts Survival in Patients with Lung and Bladder Cancer Treated with Durvalumab. Clinical Cancer Research. 2018;24(24):6212-22. doi: 10.1158/1078-0432.Ccr-18-0386.

6.         Phallen J, Leal A, Woodward BD, Forde PM, Naidoo J, Marrone KA, et al. Early Noninvasive Detection of Response to Targeted Therapy in Non–Small Cell Lung Cancer. Cancer Research. 2019;79(6):1204-13. doi: 10.1158/0008-5472.Can-18-1082.

7.         Fernandez LM, São Julião GP, Figueiredo NL, Beets GL, van der Valk MJM, Bahadoer RR, et al. Conditional recurrence-free survival of clinical complete responders managed by watch and wait after neoadjuvant chemoradiotherapy for rectal cancer in the International Watch & Wait Database: a retrospective, international, multicentre registry study. Lancet Oncol. 2021;22(1):43-50. Epub 2020/12/15. doi: 10.1016/s1470-2045(20)30557-x. PubMed PMID: 33316218.

8.        Murahashi S, Akiyoshi T, Sano T, Fukunaga Y, Noda T, Ueno M, et al. Serial circulating tumour DNA analysis for locally advanced rectal cancer treated with preoperative therapy: prediction of pathological response and postoperative recurrence. British Journal of Cancer. 2020;123(5):803-10. doi: 10.1038/s41416-020-0941-4.

9.         Slaughter DP, Southwick HW, Smejkal W. Field cancerization in oral stratified squamous epithelium; clinical implications of multicentric origin. Cancer. 1953;6(5):963-8. Epub 1953/09/01. doi: 10.1002/1097-0142(195309)6:5<963::aid-cncr2820060515>3.0.co;2-q. PubMed PMID: 13094644.

10.         Dudley JC, Schroers-Martin J, Lazzareschi DV, Shi WY, Chen SB, Esfahani MS, et al. Detection and Surveillance of Bladder Cancer Using Urine Tumor DNA. Cancer Discovery. 2019;9(4):500-9. doi: 10.1158/2159-8290.Cd-18-0825.

11.        Forde P, Spicer J, Lu S, editors. Nivolumab (NIVO)+ platinum-doublet chemotherapy (chemo) vs chemo as neoadjuvant treatment (tx) for resectable (IB-IIIA) non-small cell lung cancer (NSCLC) in the phase 3 CheckMate 816 trial. American Association for Cancer Research Annual Meeting; 2021.

12.        Magbanua MJM, Swigart LB, Wu HT, Hirst GL, Yau C, Wolf DM, et al. Circulating tumor DNA in neoadjuvant-treated breast cancer reflects response and survival. Annals of Oncology. 2021;32(2):229-39. doi: 10.1016/j.annonc.2020.11.007.

Image Credit: Aadel Chaudhuri, Angela Hirbe, Jack Shern, Jeff Szymanski, Taylor Sundby, DrawImpacts