DecisionDx-SCC Predicts Recurrence, Guides Imaging in Squamous Cell Carcinoma

Two studies were recently published that validated the use of the DecisionDx-SCC test as a tool for the treatment of cutaneous squamous cell carcinoma.

The DecisionDx-SCC test predicts local recurrence in patients with squamous cell carcinoma (SCC) deemed high-risk by NCCN guidelines who have undergone Mohs resection, and the tool’s results in guiding response to adjuvant radiation therapy (ART) and surveillance imaging have been verified via a patient survey, according to a press release from the developer, Castle Biosciences.¹

Recently, 2 new studies were published in SKIN The Journal of Cutaneous Medicine,supporting the use of the DecisionDx-SCC test. The first study, a retrospective analysis, demonstrated that 40-gene expression profile (GEP) testing stratified local recurrence-free survival (LRFS) and metastasis-free survival (MFS), and thus was a predictor of local recurrence and metastasis in patients with NCCN high-risk cutaneous SCC.² The second study, a clinician survey, demonstrated that 40-GEP test results were one of the most important factors in assessing risk of disease progression used to guide management decisions using ART and surveillance imaging.³

“These new data indicate that DecisionDx-SCC test results provide individualized risk predictions that doctors can use to guide risk-aligned escalation or de-escalation of care in their patients with NCCN high-risk SCC,” stated Désirée Ratner, MD, a Mohs micrographic surgeon and clinical professor of dermatology at the NYU Grossman School of Medicine.¹ “The ability of the test to reliably identify those patients with NCCN high-risk SCC at risk of developing local recurrence or metastasis is not only practice-changing for physicians who treat SCC, but also life-changing for their patients.”

The Retrospective Analysis

The study compiled 414 patients with NCCN high-risk cutaneous SCC who had definitive negative margins from Mohs surgery from a previously published cohort and analyzed for risk prediction of LRFS and MFS.

Of all patients, 37 (8.9%) experienced local recurrence and 25 (6.0%) experienced regional and/or distant metastases; of those who experienced local recurrence, 14 (n = 37.8%) developed regional metastasis, with all metastatic events occurring after or concurrently with recurrence. The metastatic rate in patients who did not experience local recurrence was 2.9%.

The rate of local recurrence for patients with 40-GEP class 1, 40-GEP class 2B, and 40-GEP class 2A results was 5.8%, 28.6%, and 14.5%, respectively (P = .003). The metastasis rate was 2.9%, 42.9%, and 10.7% (P <.001).

In multivariable analysis, the only significant predictors of local recurrence were 40-GEP class 2A (HR, 2.6; P = .005) and 40-GEP class 2B (HR, 6.5; P = .013).

Overall, the 3-year LRFS and MFS with 40-GEP testing were 95.3% (95% CI, 92.8%-97.8%) and 97.1% (95% CI, 95.1%-99.1%), respectively, in class 1; in class 2A, they were 85.5% (95% CI, 79.7%-91.7%) and 89.3% (95% CI, 84.2%-94.8%); and in class 2B, they were 71.4% (95% CI, 44.7%-100%) and 57.1% (95% CI, 30.1%-100%). The 40-GEP test showed significant discriminatory capacity for LRFS (P = .001) and MFS (P <.001).

In a cohort of patients with negative surgical margins after a definitive surgical approach, the 40-GEP test significantly stratified LRFS (P = .003) and MFS (P <.001).

Univariate analysis identified that 40-GEP class 2A, 40-GEP class 2B, immunosuppressed patient status, and small caliber perineural involvement (PNI) were individual risk factors significantly associated with local recurrence (all P <.05).

A multivariable analysis with 40-GEP class 2A and class 2B, immunosuppression, and small caliber PNI, all variables were significant risk predictors with HRs of 2.6, 5.3, 2.3, and 3.7, respectively (all P <.05). Modeled likelihood ratios were compared for the clinicopathologic-only (12.27) and the clinicopathologic plus 40-GEP (21.65) models and found that 40-GEP represented a significant increase in prognostic accuracy (ANOVA, P = .009).

In the trial, archival formalin-fixed, paraffin-embedded cutaneous SCC tumor tissue with clinicopathologic factors and outcome data was obtained for patients under an institutional review board-approved study protocol. Eligible patients had a documented event of local recurrence, regional or distant metastasis, or documented follow-up of at least 3 years post-diagnosis of the primary tumor without a local or metastatic event.

Samples were analyzed using 40-GEP clinical testing standard operating procedures. The trial end points were LRFS and MFS.

The Clinician Survey

A total of 244 clinicians with sufficient experience managing patients with 40-GEP test results submitted complete responses to a survey that asked questions about which classification system was their predominant method for risk assessment for cutaneous SCC tumors presenting in their practice.

Respondents were divided in their preferences for traditional risk assessment, with 32% using individual risk factors, 28% using the Brigham and Women’s Hospital (BWH) T-staging system, and 26% using some combination of staging approaches.

Additionally, 73% of the clinicians who favored NCCN guidelines considered ART for high-risk patients, and 86% considered ART for very high-risk patients. Of the clinicians who favored BWH, 32% considered ART for patients with T2a disease, 69% for T2b, and 71% for T3.

The minimum risk for metastasis at which ART was considered in patients with clear margins was 10%, which was chosen by 36% of respondents. When language was switched to “recommend”, the most common response was a 20% or higher risk of metastasis, which was chosen by 36% of respondents. Similarly, 42% of respondents said at least a 10% risk of local recurrence was required to consider ART, and when language was changed to “recommend”, it became 20% or higher, which was chosen by 41% of respondents.

When presented with a list of 22 high-risk factors and prompted about which factor alone would elevate their concern enough to discuss surveillance imaging, the most common answer was a 40-GEP class 2B test result, which was chosen by 95% of respondents; more than half also chose a 40-GEP class 2A test result. Extensive PNI and BWH stage T3 were among other options selected by more than 80% of clinicians.

The survey with questions on topics related to cutaneous SCC management was distributed to participants between March 19, 2025, and April 4, 2025, via an email invitation if they met specific criteria. Criteria were clinical credentials of MD/DO, physician’s assistant, or nurse practitioner; familiarity with 40-GEP test results, defined as at least 10 documented requests for 40-GEP testing from January 2023 to October 2024; and documented receipt of both class 1 and class 2 test results to ensure adequate familiarity with 40-GEP test result interpretation.

References

1. New evidence published supporting use of DecisionDx®-SCC test in guiding and improving treatment pathway decisions in NCCN high-risk cutaneous squamous cell carcinoma. News release. Castle Biosciences. August 25, 2025. Accessed August 26, 2025. https://tinyurl.com/4h7hcy3x
2. Ratner D, Arron ST, Kim YJ, et al. The 40-gene expression profile test identifies patients with National Comprehensive Cancer Network high-risk cutaneous squamous cell carcinoma at high risk of poor outcomes to inform management decisions. J of Skin. 2025;9(4):2426-2443. doi:10.25251/mvr2rn83
3. Shah M, Zakria D, Rogers JH, et al. Actionable risk thresholds for adjuvant radiation therapy and surveillance imaging in high-risk cutaneous squamous cell carcinoma. J of Skin. 2025;9(4):2444-2461. doi:10.25251/83v2hr07