ctDNA Assay Offers Prognostic Insights for Immunotherapy-Treated Solid Tumors

Findings from a study on the utility of ultrasensitive circulating tumor DNA (ctDNA) as a predictive biomarker for patients with advanced solid tumors undergoing immunotherapy show that whole-genome sequencing (WGS)-based ctDNA assay offered insights into treatment efficacy, patient survival, and pseudoprogression.^1,2

The assay tracks approximately 1800 tumor-specific mutations per patient, achieving detection limits as low as 1 to 3 parts per million (PPM).

Changes in ctDNA levels measured as early as 3 weeks after treatment initiation were highly predictive of long-term outcomes. A molecular response (ctDNA-mR), defined as a ≥ 30% decrease in ctDNA, strongly correlated with improved progression-free survival (PFS) and overall survival (OS).

In the retrospective cohort, patients who achieved ctDNA-mR (≥ 30% reduction) had significantly better PFS (HR, 2.69) and OS (HR, 2.23) compared with non-mR patients. In the validation cohort, the hazard ratios were 2.14 for PFS and 2.56 for OS.

Failure to achieve mR demonstrated a positive predictive value of 100% for identifying non-responders. Conversely, 100% of patients who achieved a complete or partial response by imaging had a ctDNA decrease at the pre-C2 timepoint.

Patients who achieved ctDNA clearance at any point during treatment (ctDNA-mCR) experienced significantly prolonged survival. Durable clearance (ctDNA-dmCR) for at least 180 days was associated with 100% survival within the study's observation period.

The achievement of undetectable ctDNA levels was a strong marker for deep, durable responses and prolonged survival.

Patients achieving mCR (n = 14) in the retrospective cohort had a greater than 3-fold increase in median PFS (6.4 vs 1.9 months) and a 69% 5-year survival rate. The OS HR for not achieving mCR was 6.77. ctDNA clearance was detected a median of 230 days earlier than radiologic complete response.

All the patients (n = 8) in the retrospective cohort who attained ctDNA-dmCR survived through the entire course of the study.

In patients with unconfirmed radiologic progression (iuPD) who continued immunotherapy, ctDNA dynamics served as a crucial differentiator. Decreasing ctDNA levels prior to iuPD were associated with prolonged survival, indicating a therapeutic benefit (pseudoprogression), while increasing levels signaled true disease progression.

In multivariable analyses, ctDNA-based metrics provided significant predictive value beyond standard clinicopathologic factors such as the presence of liver metastases. Incorporating ctDNA data substantially improved the accuracy of survival prediction models.

The study's findings held true across a diverse range of 24 solid tumor types and multiple immunotherapy modalities, including standard immune checkpoint inhibitors (ICIs), combination therapies, bispecific antibodies, and T-cell engagers, highlighting the universal potential of this biomarker.

Rodrigo Toledo, MD, group leader of the Biomarkers and Clonal Dynamics Laboratory at VHIO and senior author of the study, noted, "Immunotherapy has revolutionized cancer care, but response patterns can be difficult to interpret using imaging alone. Our findings show that ultrasensitive ctDNA dynamics provide a clear, early molecular view of benefit or non-response, offering a powerful tool to guide patient management across a broad spectrum of solid tumors."

Patients with Stable vs Progressive Disease

Patients with stable disease, who have highly heterogeneous outcomes, were stratified using a clustering analysis of their longitudinal ctDNA profiles. This identified 2 groups with significantly different survival outcomes, separating those with prolonged benefit from those with slowly progressing disease.

Among patients classified with progressive disease (PD) by imaging, early ctDNA dynamics identified a highest-risk subgroup. PD patients who also had an increase in ctDNA at the pre-C2 timepoint had significantly inferior OS compared to PD patients with stable or decreasing ctDNA.

Study Design and Ultrasensitive Methodology

The research evaluated the clinical utility of a next-generation, ultrasensitive ctDNA analysis across a pan-cancer population receiving various immunotherapies.

A total of 202 patients with metastatic (stage IV) solid tumors were enrolled from a phase 1 clinical trial unit. A total of 136 patients were in the retrospective cohort, and a total of 66 patients in the validation cohort.

The study included 24 different cancer types which were grouped into 11 categories for analysis, including breast, colorectal, non-small cell lung, and melanoma.

Patients were treated with a range of cancer immunotherapies, including ICI monotherapies (anti-PD-1, anti-PD-L1, anti-CTLA-4, LAG-3), ICI combination therapies, and emerging modalities like bispecific antibodies and T-cell engagers.

A total of 1455 longitudinal plasma samples were analyzed, with an average of 7.2 samples per patient, enabling detailed tracking of ctDNA dynamics over time.

The study utilized a tumor-informed WGS-based assay (NeXT Personal) to achieve high sensitivity and specificity. WGS was performed on both tumor tissue and a matched normal (blood) sample for each patient.

A bespoke panel of up to 1800 patient-specific somatic mutations was designed for tracking in plasma. The median number of variants selected per patient was 1830. The personalized panel was used to perform deep sequencing on cell-free DNA (cfDNA) extracted from longitudinal plasma samples to quantify ctDNA levels.

The assay achieved a median limit of detection of 1.9 PPM. ctDNA was detected in 98.0% of patients at baseline before treatment initiation. Of the on-treatment ctDNA detections, 11.1% occurred in the ultrasensitive range (< 100 PPM), underscoring the necessity of high-sensitivity assays.

Differentiating True Progression from Pseudoprogression

Immunotherapy can cause pseudoprogression, where tumors appear larger on scans due to immune infiltration. The study analyzed 45 patients who continued treatment after an initial iuPD event.

The trajectory of ctDNA at the time of iuPD was highly predictive of the true disease state.

Decreasing ctDNAwasassociated with significantly improved survival (1-year survival of 80% vs 39% in the retrospective cohort). This indicates the patient was likely experiencing pseudoprogression and benefiting from continued therapy.

Increasing ctDNA indicated true disease progression and a poor prognosis for continued treatment.

The Role of Ultrasensitivity

The study demonstrated that the assay's PPM-level sensitivity was essential for accurate risk stratification. A simulation was performed by treating all ctDNA levels below 100 PPM as "undetected," mimicking a less sensitive assay. This resulted in diminished predictive power and incorrectly assigned patients with minimal residual disease to the "clearance" group, thereby weakening the correlation with survival.

The analysis was performed retrospectively, and the tumor-informed approach has a longer turnaround time and requires a tumor sample, which may not always be feasible.

REFERENCES

1.Garralda E, Abbott C, Calahorro A, et al. Broad utility of ultrasensitive analysis of ctDNA dynamics across solid tumors treated with immunotherapy. Clin Cancer Res OF1–OF17. December 16, 2025. doi: 10.1158/1078-0432.CCR-25-2312.

2.Personalis announces new publication applying ultrasensitive ctDNA testing to monitoring cancer immunotherapy response across solid tumors. News release. Personalis. December 17, 2025. Accessed December 18, 2025. https://tinyurl.com/2sz9yehv