24 Utility of tumor-informed molecular residual disease assays in patients with complete response to immune checkpoint blockade

Abstract

BackgroundRecent data suggests that responses in patients with mismatch repair deficient (dMMR) tumors tend to be durable and potentially curative. Typically immunotherapy is employed and approved only in metastatic settings. However, it is not uncommon now to consider usage in patients with dMMR tumors secondary to the hypermutated nature of these malignancies alongside the concerns that these do not respond to therapy either through a clinical trial or off-label compassionate access programs. As noted, in patients who have these dramatic and durable responses, the question of foregoing surgery and/or radiation comes up. There is no great test to help predict or guide who are these complete responders. Assessment of molecular residual disease or minimal residual disease through tumor-informed assays is one potential test that can be employed in this setting. We here show the feasibility of such an approach in patients with dMMR tumors who got immunotherapy in the advanced but not metastatic setting.MethodsWe identified patients who were enrolled in the serial tumor informed molecular residual disease BESPOKE expanded access ctDNA testing program (Signatera 16-plex bespoke mPCR NGS assay) who got immunotherapy in the neoadjuvant setting and were also enrolled in our biobanking program.ResultsWe were able to serially do ctDNA analysis in 2 patients (1 with advanced but not metastatic esophageal adenocarcinoma and another patient with advanced but not metastatic nearly obstructing rectal adenocarcinoma with extensive nodal metastases in both situations) who got immunotherapy off-label per physician discretion and tumor board discussion. Of note both these patients also had germline lynch syndrome. Both of them had robust, dramatic and ongoing responses to immune checkpoint blockade. Of note, while they both had some radiographic question of residual disease, repeated endoscopic ultrasounds and random biopsies have yielded no tumor and all just scarred tissue. Circulating tumor DNA in these instances quickly declined to become undetectable and has remained not detectable to date (figure 1). Surgery and radiation is deferred and close follow up alongside serial endoscopic/radiographic assessment in addition to novel usage of ctDNA MRD assays is being employed.Abstract 24 Figure 1Ongoing response and negative ctDNA minimal residual disease to PD-1 blockade added to standard of care chemotherapy for a patient with advanced rectal cancerConclusionsData regarding neoadjuvant usage of immunotherapy is scarce pertaining to mismatch repair deficient tumors. There is a case series of 3 patients with rectal cancer that was just reported this year.1 Here we not only report a case of an advanced rectal cancer, but also a case of an advanced esophageal cancer who have achieved dramatic responses with no evidence of disease on repeated sampling. It is rare for surgery or radiation to be deferred in these situations. However, with added utility of ctDNA MRD assays, it gives us one more tool in our toolbox in terms of applying it to patients suitable for these ‘watch-&-wait’ approaches.Ethics ApprovalThe study has been approved by University of Iowa’s Institutional Review Board IRB#201202743.ConsentWritten informed consent was obtained from the patient for publication of this abstract and any accompanying images. A copy of the written consent is available for review by the Editor of this journal.ReferenceDemisse R, Damle N, Kim E, et al. Neoadjuvant Immunotherapy-Based Systemic Treatment in MMR-Deficient or MSI-High Rectal Cancer: Case Series. J Natl Compr Canc Netw. 2020;18(7):798–804. doi:10.6004/jnccn.2020.7558

Liquid biopsies for residual disease and recurrence

“…6,29,78 RDR testing may also have utility following neoadjuvant immunotherapy in patients with mismatch repairdeficient CRC, where patients whose ctDNA became undetectable may be able to defer surgery or radiation. 79 Initial RDR assays targeted single loci with known mutations or gene panels of limited breadth and have subsequently broadened to cover multiple markers present in the tumor, through sequencing of larger fixed panels of genes or methylation sites or by deep sequencing of patient-specific mutation panels (methods are discussed in Box 2 and Figure 2 and are listed in Table 1). Regardless of the technical approach taken, assays for RDR require sensitivity and specificity: high positive predictive value (PPV) would be important for treatment escalation decision-making and high negative PV (NPV) for treatment de-escalation.…”