Supplementary Material For:The Radiogenomic Risk Score Stratifies Outcomes in a Renal

Supplementary Material for:The Radiogenomic Risk Score stratifies outcomes in a renal cell cancer phase 2 clinical trial

Figure E1: Illustrative example of a hypothetical scenario. Consider two hypothetical patients with two similar appearing tumors at an initial time point (‘o’ on the phenotypic risk map), thus placing both people in the same risk category based on the imaging appearance of ccRCC. At a later time point when the patients are re-imaged, bother tumors have altered appearances, corresponding to ‘x’ and ‘v’ for hypothetical patients 1 and 2, respectively. The risk, as assessed by the appearance of the tumor and the RRS has not changed for patient 1, however it has changed for patient 2.

Figure E2: Set of Kaplan Meier survival plots for WHO performance, stage, and grade. The 0 and 1 curves represent the low and high RRS groups, respectively, in each subgroup analysis. The RRS outperforms these and is independent of each of these measures except in lower stage (stage 1 and 2) tumors.

Table E1

Pearson correlation coefficients between the different images traits which comprise the RRS. The rank of the imaging feature matrices was 4 (for the training and validation data sets), i.e. all of the imaging traits are linearly independent of one another.

Table E2

Gene Ontology enrichment of the RRS. The comprehensive list of GO term enrichment based upon the set of genes that were significantly correlated with the radiogenomic risk score from the training and validation sets (p < 0.05).

Description of Clinical Trial inclusion and exclusion criteria

As described in Jonasch et al [1], the bevacizumab study was a single center study requiring,

1) histologically or cytologically confirmed metastatic clear cell renal cell carcinoma,

2) primary tumor deemed resectable by the treating urologist,

3) measurable disease by Response Evaluation Criteria in Solid Tumors (RECIST) not including primary tumor,

4) Eastern Cooperative Oncology Group performance status of 0 or 1,

5) adequate organ and marrow function within 14 days, including absolute neutrophil count >=? 1,500/?L, platelets >= 75,000/microL, hemoglobin > 9.0 g/d (transfusion allowed), total bilirubin ?<= 2.0 mg/dL, albumin > 3.0 g/dL, serum creatinine <= 2.0 mg/dL, AST and/or ALT <= 2.5 times the institutional upper limit of normal for participants without evidence of liver metastases, and AST and/or ALT <= five times institutional upper limit of normal for participants with documented liver metastases,

6) absence of brain metastases,

7) no prior systemic therapy

The laboratory criteria are in place to ensure relative medical stability (i.e. no acute or chronic renal or hepatic disease) to avoid any confounding factors that may contribute to a potential lack of response.

The imaging features were interpreted on the pre-contrast and equilibrium phase (70 second post contrast) acquisitions and were assessed for 3 consecutive slices spanning the largest transaxial diameter of the tumours. Further details are provided in Jamshidi et al [2].

1Jonasch E, Wood CG, Matin SF et al (2009) Phase II presurgical feasibility study of bevacizumab in untreated patients with metastatic renal cell carcinoma. J Clin Oncol 27:4076-4081

2Jamshidi N, Jonasch E, Zapala M et al (2015) The Radiogenomic Risk Score: Construction of a Prognostic Quantitative, Noninvasive Image-based Molecular Assay for Renal Cell Carcinoma. Radiology 277:114-123