- Clinical Development        - Discovery Background        - Additional Development and Validation        - Laboratory Report        - Publication List     - Reimbursement

Discovery Background

DecisionDx-UM is a robust, proprietary, multi-gene expression test that analyzes the molecular signature of your patient’s uveal melanoma. The DecisionDx-UM molecular signature enables prospective identification of patients who have a low risk (Class 1 molecular signature) or high risk (Class 2 molecular signature) of developing metastatic disease.

The DecisionDx-UM assay was discovered at Washington University in St. Louis in the laboratory of Dr. J. William Harbour.

Validation is complete and this assay is now available for clinical use through Castle’s CLIA-certified laboratory.

The following information provides background on the assay development and validation studies. Further information can be provided upon request.

As shown in Table 1 below, a number of development considerations have been addressed during the development phase of the DecisionDx-UM assay.

Table 1: DecisionDx-UM Development Considerations
Development Considerations	Solution
Identification of robust target gene signature;  addressing potential errors from overfitting,  redundancy, and instability	Conducted studies identifying gene targets and signatures from multiple independent sample sets, multiple analytical layers, and multiple platforms.
Clinically appropriate specimen collection methodology	Studies confirmed functionality of assay with specimens through fine-needle aspirate biopsies (FNAB) prior to brachytherapy and immediately post-enucleation. Studies also confirmed functionality in formalin-fixed paraffin embedded enucleated globes.
Comparison to other prognostic markers	Study was undertaken to compare DecisionDx-UM to the well-known chromosomal marker monosomy 3 as well as clinical and pathologic factors currently used for prognosis. DecisionDx-UM was superior to monosomy 3 and clinicopathologic prognostic factors at predicting metastasis.
Validation in separate retrospective and  prospective studies	DecisionDx-UM has been validated in published retrospective studies and in a large (n=650), multi-center prospective study.

Comparison to Other Prognostic Markers (Worley, et al 2007)

A study was undertaken to compare the gene expression profile (molecular signature) to the chromosomal marker - monosomy 3. The same study compared the assay to clinical and pathologic factors for predicting metastasis in uveal melanoma. This study of 67 patients treated by enucleation is the largest study using metastatic outcome to compare prognostic factors to date in uveal melanoma. Monosomy 3 was assessed through fluorescence in situ hybridization (FISH) and array comparative genomic hybridization (aCGH). The clinicopathologic factors of age, sclera invasion, histopathologic cell type, tumor thickness, gender, largest tumor diameter, and anterior tumor location were also assessed.

By Cox univariate proportional hazards, only Class 2 gene expression profile (P = 0.0001), advanced patient age (P = 0.01), and scleral invasion (P = 0.007) were significantly associated with metastasis (Table 2). Kaplan-Meier analysis rendered similar results. When all three significant variables were entered into a Cox multivariate model, only the Class 2 molecular signature exhibited significant association with metastasis.

Table 2: Extracted Summary (Additional clinicopathologic factors eliminated due to non-significant outcomes)
Prognostic Factor	Cox univariate	Cox multivariate	Kaplan-Meier
Class 2 molecular signature	0.0001	0.0449	0.0001
Monosomy 3 (aCGH)	0.0762	-	0.0856
Monosomy 3 (FISH)	0.1275	-	0.182
Age	0.0144	0.4703	0.007

For sensitivity and specificity analyses, a metastasis group consisted of 18 patients who had developed metastasis, and a non-metastasis group contained 16 patients without metastasis who had at least 3-year follow-up after ocular treatment was selected. Consistent with the statistical outcomes above, the Class 2 molecular signature outperformed all other prognostic variables (Table 3).

Table 3: Extracted Summary (Additional clinicopathologic factors eliminated due to non-significant outcomes noted above)
Prognostic Factor	Sensitivity	Specificity	Positive  likelihood  ratio	Negative  likelihood  ratio	Positive  predictive  value	Negative  predictive  value
Class 2 molecular signature	85%	93%	12%	0.2%	92%	87%
Monosomy 3 (aCGH)	58%	86%	4%	0.5%	78%	71%
Monosomy 3 (FISH)	50%	73%	2%	0.7%	63%	62%
Age	72%	69%	2%	0.4%	72%	69%

[back to top]

---

Prospective Validation

The DecisionDx-UM molecular signature has been evaluated in over 700 patients with uveal melanoma to date. Of these, over 650 patients have participated in a prospective, multi-center, 5 year follow-up study to validate the predictive accuracy of the gene expression-based molecular test. Outcomes are collected and the ability of the molecular diagnosis to predict metastasis is being evaluated at regular intervals.

At the most recent collection point (November 3, 2009) the prospective study included 227 patients with evaluable data for analysis. The actual outcomes for metastasis and survival of the predicted low-risk (Class 1) molecular signature and the high-risk (Class 2) molecular signature with a 72 months analysis cut-off are shown below.

[back to top]

---

Laboratory Report

A laboratory report will be generated and provided to each ordering and treating clinician for each successful performance of a clinical order. The report will be available through facsimile, mail, and secure web portal.

The laboratory report will contain the predicted classification (Class 1 vs Class 2) for the analyzed tumor specimen as well as the corresponding discriminant value.

[back to top]

---

Publication List

AJCC Cancer Staging Manual. 2010. Edge, S.B.; Byrd, D.R.; Compton, C.C.; Fritz, A.G.; Greene, F.L.; Trotti, A. (Eds.) 7th ed.

 Onken, et al. 2004. Gene expression profiling in uveal melanoma reveals two molecular classes and predicts metastatic death. Cancer Res 2004;64: 7205-9.

 Onken, et al. 2006. Prognostic testing in uveal melanoma by transcriptomic profiling of fine needle biopsy specimens. Jrnl Mol Diagn; 8: 567-73.

 Worley, et al. 2007. Transcriptomic versus Chromosomal Prognostic Markers and Clinical Outcome in Uveal Melanoma Clin Cancer Res 2007;13(5): 1466-71.

[back to top]

---

Reimbursement

The DecisionDx-UM assay is a proprietary assay that is only available through Castle Biosciences.

The DecisionDx-UM assay stratifies patients who have a near term (5-year) low risk versus a high risk of developing metastatic disease – regardless of the treatment effectiveness of their primary uveal melanoma tumor.

Based upon the completion of the validation of this gene expression assay and its clinical importance in the management of patients with uveal melanoma, the 7th Edition of the American Joint Committee on Cancer (effective January 1, 2010) has identified this assay as "key prognostic factor that (is) important to collect in [uveal melanoma]". The DecisionDx-UM gene expression assay is recommended for collection because it was determined to be "clinically significant".

Based upon the recommendation of the AJCC, we anticipate that insurance companies will reimburse for the cost of the assay.

However, we understand that paying for elements of healthcare can be confusing and overwhelming. As a company, we are sensitive to that fact and are prepared to assist patients through the process of securing payment for services provided by Castle. We will coordinate the submission of insurance claims for Medicare and commercial insurance companies. We will work with the patient's authorized healthcare provider regarding benefits investigation should one be requested, any prior authorization required, and appeals processing. In cases when, after all appeals are exhausted, the claim is not fully paid then the patient will ultimately be responsible for any balance remaining.

The Castle Patient Assistance Program is available to explore options for patients with balances remaining. This program also has options for uninsured patients.

In some cases, institutions may ask us to invoice them directly, also known as client bill.

[back to top]