ctDNA Shows Promise as a Potential Biomarker for MRD in Locally Advanced NSCLC

Maximilian Diehn, MD, PhD, discusses how the use of circulating tumor DNA to analyze minimal residual disease status in patients with lung cancer and other solid tumors is a rapidly developing field with the potential to further personalize treatment decisions about adjuvant therapy.

The use of circulating tumor DNA (ctDNA) to analyze minimal residual disease (MRD) status in patients with lung cancer and other solid tumors is a rapidly developing field with the potential to further personalize treatment decisions about adjuvant therapy, according to Maximilian Diehn, MD, PhD.

In non–small cell lung cancer (NSCLC), ctDNA is strongly prognostic among patients with localized disease and may help inform recommendations about whether consolidation therapy should be administered after surgery, Diehn said during a live virtual presentation at the 15th Annual New York Lung Cancers Symposium®.1

“This is one of the emerging applications of ctDNA that goes beyond just using this analyte for noninvasive tumor genotyping,” said Diehn, vice chair of research and division chief of radiation and cancer biology at Stanford Cancer Institute and the Institute for Stem Cell Biology & Regenerative Medicine, both at Stanford University in California. “I think it’s one that is very close on the horizon that will transform how we manage [disease in] our patients.”1

Just as cell-free DNA leaks into the circulation as part of healthy biological functions, tumors also shed DNA that is mixed in with the patient’s germline DNA, said Diehn, who also is the CRK Faculty Scholar and an associate professor at Stanford. The ctDNA can be detected in blood plasma or serum samples, much like identifying tumor cells in tissue. “You can think of a blood draw from a patient as a very dilute tumor biopsy,” said Diehn.

Mutation-based assays for the detection ctDNA can be used in either tumor-naïve or tumor-informed settings, Diehn said. Tumor-informed tests, designed for surveillance among patients known to have a mutation, have the highest analytical sensitivity of approximately 0.002% whereas the tumor-naïve assays, used primarily for genotyping and early detection in people without prior knowledge of a mutation, have a detection limit of about 0.2%. An emerging ctDNA method that uses a methylation-based approach for tumor-naïve analysis also has a sensitivity of about 0.2%.1

As it stands now, the best option for MRD analysis is a mutation-based assay that uses a tumor-informed approach, according to Diehn. During the past several years, Diehn and colleagues have developed a next-generation sequencing-based ctDNA detection method called Cancer Personalized Profiling by Deep Sequencing (CAPP-Seq). When used for MRD analysis, CAPP-Seq can be used to genotype tumor using tissue from patients with stage I-III disease or pretreatment plasma from stage III, and to sequence posttreatment plasma to determine MRD status. Diehn said the assay has a sensitivity of 0.002%.

In illustrating the potential utility of ctDNA as a biomarker in NSCLC, Diehn noted findings from a study in 85 patients with localized disease that used CAPP-Seq technology. Investigators found that ctDNA was detected in 42%, 67%, and 88% of patients with stage I, II, and III disease, respectively, but the concentration of DNA shed by stage was low, particularly in stage I. The ctDNA concentration was higher in patients with more tumor volume. Additionally, the detection rate differed by histology, with 43% of adenocarcinoma versus 95% of non-adenocarcinoma samples with ctDNA.2

Although investigators have not yet been able to develop MRD assays for most solid cancers, Diehn said the prospects of being able to do so hold great potential. “ctDNA appears to be a very promising approach for detecting minimal residual disease in a variety of solid tumors, including lung cancers,” he said. “If we could detect which patients have MRD after surgery for an early-stage lung cancer, we could potentially use that information to personalize further treatment. One could imagine that a patient who has no detectable ctDNA after surgery is likely cured and therefore may have little chance of benefiting from adjuvant therapy whereas a patient who has detectable ctDNA is almost assuredly not cured by the surgeon because there’s still evidence of tumor left and those patients might be the ideal group to which we would want to offer adjuvant therapy.”

The question of whether ctDNA can be used as a biomarker for MRD in this setting has generated considerable debate, Diehn said. Some experts, he said, argue that ctDNA analyses fail to detect residual disease in many patients and that those who are ctDNA-positive have too much disease burden to benefit from the testing whereas other investigators say the assays are sufficiently sensitive and that some patients who are MRD-positive using these methods can gain from treatment.

Randomized trials to evaluate these competing hypotheses are getting started “but it will be years before they read out,” Diehn said. In the meantime, investigators studied the impact of ctDNA in a clinical scenario in stage III NSCLC in light of findings in the landmark phase 3 PACIFIC trial (NCT02125461). The study evaluated the PD-L1 inhibitor durvalumab (Imfinzi) versus placebo in patients with stage III unresectable NSCLC who did not have disease progression after concurrent chemoradiotherapy (CRT).3 Progression-free survival (PFS) findings from the study prompted the FDA to approve durvalumab in this patient population in February 2018.4

In updated findings for the overall study population, durvalumab therapy resulted in a median PFS of 17.2 months (95% CI, 13.1-23.9) in compared with 5.6 months (95% CI, 4.6 -7.7) in the placebo group, which translated into a 49% reduction in the risk disease progression or death (HR, 0.51; 95% CI, 0.41- 0.63).3 However, Diehn said PFS curves indicate that although some patients were cured by durvalumab therapy and others were cured by CRT without further treatment, the largest group of patients did not derive benefit from consolidation therapy.1

In order to test whether ctDNA could predict which patients would benefit from adjuvant immunotherapy in this setting, Diehn and colleagues conducted a retrospective study of 62 patients with stage III NSCLC: 37 who received CRT without further treatment and 25 who received CRT followed by consolidation with an immune checkpoint inhibitor (ICI).5

Patients with negative ctDNA after CRT had better outcomes and similar responses whether or not they received consolidation with an ICI, Diehn said, suggesting that this group of patients would not benefit from further treatment. Moreover, 1 patient with detectable ctDNA prior to CRT who was ctDNA-negative after CRT and received an ICI went on to develop grade 5 drug-induced pneumonitis. Diehn said the autopsy showed no residual cancer.1,5Analyzing ctDNA, he said, could provide clues about whom not to treat with adjuvant therapy.

Meanwhile, patients without a ctDNA response during consolidation ICI therapy also did not seem to benefit from continued treatment, suggesting they had developed intrinsic resistance, Diehn said. The “most exciting” outcomes were among patients who had detectable ctDNA after CRT but achieved negative ctDNA after ICI therapy. “These are the ones who are really deriving benefit from consolidation immunotherapy,” he said.

Although Diehn focused on findings from his research group during the presentation, he said other investigators using different assays had similarly demonstrated that ctDNA MRD has strong prognostic power in studies involving more than 200 patients with localized NSCLC.

Looking forward, a signal of potential clinical adoption of ctDNA assays for MRD assessment came in October 2020 when the Centers for Medicare & Medicaid Services issued a proposed local coverage determination (LCD) for validated tests able to detect molecular recurrence or progression before it is evident through clinical or radiographical evaluation. The LCD, which is open for comment until November 22, 2020, would cover patients with advanced cancer or a personal history of cancer, including solid tumors and hematologic malignancies.6

“I believe the glass is half-full,” Diehn said. “We can use ctDNA MRD to identify patients who might be the best candidates for consolidation or adjuvant therapy and this would allow us to personalize additional therapy in the early stage patients in a way we are currently not doing.”


  1. Diehn M. The emerging role of ctDNA as a biomarker for MRD and treatment decision-making in NSCLC. Presented at: 15th Annual New York Lung Cancers Symposium®. Virtual. November 7, 2020.
  2. Chabon JJ, Hamilton EG, Kurtz DM, et al. Integrating genomic features for non-invasive early lung cancer detection. Nature. 2020;580(7802):245-251. doi:10.1038/s41586-020-2140-0
  3. Antonia SJ, Villegas A, Daniel D, et al; PACIFIC Investigators. Overall survival with durvalumab after chemoradiotherapy in stage III NSCLC. N Engl J Med. 2018;13;379(24):2342-2350. doi:10.1056/NEJMoa1809697
  4. FDA approves durvalumab after chemoradiation for unresectable stage III NSCLC. FDA/ February 20, 2018. Accessed November 9, 2020. https://bit.ly/3n6gHc3
  5. Moding EJ, Liu Y, Nabet BY, et al. Circulating tumor DNA dynamics predict benefit from consolidation immunotherapy in locally advanced non-small-cell lung cancer. Nat Cancer. 2020; 1:176-183. doi:10.1038/s43018-019-0011-0
  6. Proposed local coverage determination (LCD): MolDX: minimal residual disease testing for cancer (DL38816). Centers for Medicare & Medicaid Services. October 1, 2020. Accessed November 9, 2020. https://go.cms.gov/38uTe03