PowerPoint Presentation


Improved prognostic guidance by the 31-gene expression profile test for clinical decisions after a negative lymph node for 
patients with cutaneous melanoma

Background
›Despite a good overall prognosis for patients with a negative
SLNB, 10-24% will experience recurrence or metastasis, and
melanoma-specific survival (MSS) rates range from 82-99%.1-4 A
subset of these patients (stage IIB-IIC) are currently eligible for
adjuvant therapy, though it is unclear which patients will benefit and
which patients do not need therapy.5

›The recent KEYNOTE-716 trial showed a benefit of adjuvant
pembrolizumab in patients with stage IIB-IIC melanoma (9% RFS
improvement), but 80% had an adverse event (16% grade 3 and
higher), and 18% discontinued treatment due to adverse events.5

›These data underpin a need for prognostic tools beyond
clinicopathologic features to identify patients with high-risk
tumor staging but low-risk tumor biology, or low-risk tumor
staging but high-risk tumor biology, so that patients receive
risk-aligned treatment.1-2

›Multiple prospective and independent studies have shown that the
31-GEP test is a consistent and independent predictor of survival
outcomes in large populations of patients with stage I-III CM, and
that clinicians use the 31-GEP to guide patient management
decisions.3, 6-10

Acknowledgments & Disclosures

References

›SK, CB, BM, MG, RC, and KC are employees and shareholders of Castle Biosciences, Inc.
›VP has no disclosures

Results

Presented at the Winter Clinical Dermatology Conferences 2023.

Brian Martin PhD1, Christine Bailey, MPH1, Matthew Goldberg, MD1,2, Valentina Petkov, MD, MPH3, Robert Cook, PhD1, Kyle Covington, PhD1, Sarah Kurley, PhD1

1Castle Biosciences, Inc., Friendswood, TX 2Icahn School of Medicine at Mount Sinai, New York, NY, 3National Cancer Institute, Surveillance Research Program, Bethesda, MD

For more information: mgoldberg@castlebiosciences.com

Objective
In collaboration with the National Cancer Institute’s Surveillance,
Epidemiology, and End Results (SEER) program (covering 34% of
the U.S. population during the study period) this study sought to:
›Demonstrate the performance of the 31-GEP to identify
patients with high-risk tumor biology in an unselected,
clinically tested cohort of patients with a negative lymph node.

›In patients with a negative lymph node, the 31-GEP
identifies patients more or less likely to die from their
melanoma in the absence of adjuvant therapy, and the
31-GEP is a significant predictor of melanoma-specific
death, even when accounting for substage.
›The 31-GEP can direct care to patients with high-risk
tumor biology who are most likely to benefit from higher
intensity management and away from those unlikely to
benefit from adjuvant therapies to spare patients from
adjuvant therapy-associated adverse events.

Conclusions

›SEER cancer registries linked CM cases diagnosed from 2016-2018 to data
for patients with CM who were tested with the 31-GEP (n=3,271). Linkage
was mediated by Information Management Services (an Honest Broker for
the SEER registries). A de-identified dataset was used for this analysis. A
focused analysis of negative lymph node patients was performed.

›Kaplan-Meier analysis with the log-rank test was used to analyze 3-year
melanoma-specific survival (MSS). Multivariable Cox regression was used to
identify factors associated with MSS.

Scan or click here  for 
more info

Methods

1. Thomas, D. C. et al. Ann Surg Oncol 26, 2254–2262 (2019). 2.Jones, E. L. et al. JAMA surgery 148, 456–61 (2013). 3. Greenhaw, et al. Dermatol Surg 44, 

1494–1500 (2018). 4. O’Connell, E. et al. Mel Research 26, 66-70 (2016). 5. Luke et al. Lancet, 399, 1718-1729 (2022) 6. Hsueh, E. C. et al. JCO Precision 

Oncology 5, 589–601 (2021). 7. Vetto, J. T. et al. Future Oncology 15, 1207–1217 (2019). 8. Podlipnik, S. et al. J Eur Acad Dermatol Venereol 33, 857–862 

(2019). 9. Dillon, L. D. et al. SKIN J Cutaneous Med 2, 111–121 (2018). 10. Berger, A. C. et al. Curr Med Res Opin 32, 1599–1604 (2016).

Figure 1. Using the 31-GEP in patients with a negative 
lymph node identifies those at highest risk of dying from 

their disease.

Table 1. Multivariable analysis demonstrates independent 
and significant prognostic information compared to 
traditional staging factors

Melanoma-specific survival Multivariable HR (95% CI)

31-GEP Class 1A Reference

31-GEP Class 1B/2A 5.76 (1.42-23.41)

31-GEP Class 2B 10.50 (2.55-43.28)

Age (continuous) 1.05 (1.02-1.08)

AJCC Stage IA Reference

AJCC Stage IB 1.48 (0.37-6.01)

AJCC Stage IIA 3.93 (1.10-14.12)

AJCC IIB 3.24 (0.82-12.86)

AJCC IIC 4.58 (1.09-19.22)

Clinical Impact

›Patients with Class 1A results had higher 3-year MSS (Class 1A:
99.7%; Class 1B/2A: 97.8%; Class 2B: 91.8%, p<0.001).
› In the subset of patients with IIB-IIC disease (n=311), no Class
1A (0%, 0/38) patients died from melanoma compared with
6.7% (14/210; 8 IIB, 6 IIC) of Class 2B patients.

›Using the 31-GEP results to guide increased clinical
management and surveillance for patients at high risk of
melanoma-specific death may improve patient
outcomes.

Node Neg

Class 1A

Class 

1B-2B

Continue low
intensity 

management

Consider increased intensity 
management, adjuvant therapy 

(Stage IIB, IIC and III) or clinical trials

Increased Surveillance = 
Early detection of metastasis 

(low tumor burden)

Demonstrated 

improved response to 

surgical and systemic 

therapy

https://castlebiosciences.com/research-development/publications/

	Slide 1