Clinical Summary 

• Validated Clinical Benefits (Two Validation trials: QUASAR and CALGB 9581)
• QUASAR – Largest Randomized Study of Observation versus Adjuvant 5FU/LV in Stage II Colon Cancer Now published in the Journal of Clinical Oncology
• A Quantitative Multi-Gene RT-PCR Assay for Prediction of Recurrence in Stage II Colon Cancer: QUASAR Validation Study
• Population Distribution and Recurrence Risk by Recurrence Score in the Clinical Validation Study
• Putting the Recurrence Score into Context: Independent Value Beyond Existing Clinical and Pathological Features
• Consistency of the Recurrence Score Across All Clinical Endpoints
• The Need for Individualized Risk Assessment: Limitations with Existing Clinical and Pathologic Markers
• T-Stage and Outcome in Stage II Colon Cancer – SEER database 1991-2000
• MMR/MSI and Outcome in Resected Colon Cancer
• QUASAR Study: Insight Beyond Existing Markers  

Validated Clinical Benefits of Oncotype DX

Based on data from the nearly 3,300 patients in development and validation studies, the Oncotype DX Colon Cancer Assay delivers clinically valuable information that can be used to help make decisions about adjuvant therapy. These studies, as well as data from a rigorous process of analytical validation, support the consistency, reproducibility, and accuracy of the Oncotype DX assay. The assay provides an individualized colon cancer Recurrence Score, provides member-specific information to help decide which of your stage II colon cancer members should be most appropriately considered for post-surgical chemotherapy.

The Recurrence Score is based upon the quantitative expression of the 7 cancer genes, normalized to the 5 reference genes. The colon cancer Recurrence Score includes 7 genes that have been identified as being consistently and significantly associated with recurrence free interval (RFI) in the 1,851 patients from the development studies. These genes include the cell cycle group (Ki-67, MYBL2, C-MYC), the stromal group (FAP, INHBA, BGN) and GADD45B. The pre-specified Recurrence Score gene panel was validated in 1,436 stage II colon cancer patients with tissue from the QUASAR trial.

QUASAR – Largest Randomized Study of Observation versus Adjuvant 5FU/LV in Stage II Colon Cancer 

• 2,146 patients with stage II colon cancer participated in this landmark randomized study to determine the benefit of adjuvant 5FU/LV chemotherapy compared to surgery alone
• Total QUASAR enrollment (n=3,239) included stage III and rectal cancer patients
• 5FU/LV produced a 20% relative risk reduction for both recurrence and survival in stage II colon cancer compared to surgery alone, with an absolute benefit of approximately 3-4%
• Tumor specimens were available for nearly 70% of stage II colon cancer patients in QUASAR  

A Quantitative Multi-Gene RT-PCR Assay for Prediction of Recurrence in Stage II Colon Cancer: QUASAR Validation Study

The QUASAR study results strongly support a new paradigm for assessing risk of recurrence in stage II colon cancer, emphasizing the critical role of three measures: Recurrence Score (RS), Mismatch Repair/Microsatellite Instability (MMR/MSI) and T-stage.  

• QUASAR validated the Oncotype DX Colon Cancer Assay in 1436 stage II colon cancer patients using prospectively defined endpoints including recurrence-free interval (RFI), disease-free survival (DFS), and overall survival (OS).
• In the primary analysis, the continuous Recurrence Score was significantly associated with recurrence risk (P=0.004), with a near linear relationship between RS and risk of recurrence. 3 year colon cancer recurrence risk ranged from 9-11% at low RS to 25-27% at high RS. 
• In a pre-specified multivariate analysis, Recurrence Score (P=0.008) demonstrated a significant and independent ability to predict recurrence, even after simultaneously accounting for the prognostic effects of tumor grade, lymphovascular invasion, number of nodes examined, T-stage, and MMR status. Recurrence Score, T stage and MMR status, were the most important, independent predictors of recurrence.  

Population Distribution and Colon Cancer Recurrence Risk by Recurrence Score in the Clinical Validation Study

• In a secondary analysis, recurrence risk was analyzed for 3 groups of patients defined by pre-specified cutpoints in RS. Cutpoints in RS were defined to capture sufficiently large numbers of patients in each recurrence risk group to permit statistical analyses: the low recurrence risk group included patients with RS < 30 (43.7% of patients), the intermediate recurrence risk group included patients with RS 30 to 40 (30.7% of patients), and the high recurrence risk group included patients with RS of 41 or greater (25.6% of patients). This grouping permitted a Kaplan-Meier analysis of these 3 groups for recurrence, demonstrating that the low group had a 3-year recurrence of 12% as compared with the high group, which had a 3-year recurrence of 22%.
• A protocol-specified comparison of the high versus low recurrence risk groups in a Cox model shows significant HR 1.47 (with p<0.05)
• While this grouping enabled the statistical analyses described here, the greatest clinical advantage for the Recurrence Score will be as a continuous measure, where quantitative, individual recurrence risk is incorporated with the specific clinical features of each member to make an informed, personalized colon cancer treatment plan.

Top

Putting the Colon Cancer Recurrence Score into Context: Independent Value Beyond Existing Clinical and Pathological Features

In a protocol-specified multivariate analysis, the Recurrence Score was shown to be a powerful, independent predictor of recurrence risk providing value beyond available measures. Three factors -- Recurrence Score, MMR status, and T stage -- were found to be the most significant independent predictors of recurrence risk following surgery.  

• Recurrence Score remained significantly associated with recurrence free interval (RFI) (P=0.008) after simultaneously controlling for the prognostic effects of MMR, T stage, tumor grade, number of nodes examined, and LVI, with Recurrence Score retaining the HR and significance level observed in univariate analyses.
• The 13% of patients with MMR deficiency (MMR-D) have a significantly lower recurrence risk than the 87% of patients with MMR proficiency (MMR-P). Given the small proportion of patients with MMR-D, patients with MMR-P have near-average recurrence risk. Thus, although MMR-D is a marker of low risk, MMR-P is not a marker of high risk.
• The 15% of patients with T4 stage have a significantly higher recurrence risk than the 85% of patients with T3 disease. Given the small proportion of patients with T4 stage, patients with T3 disease have similar risk to the overall stage II colon cancer population. Thus, although T4 stage is a marker of high risk, T3 stage is not a marker of low risk.
• MMR-D and T4 stage are informative for a small fraction of patients (13% and 15%, respectively), defining populations with significantly lower risk and higher risk, respectively. However, given the large proportion of members who do not have MMR-D or T4 stage, MMR and T stage are not informative for the majority of members.
• In the multivariate analysis, other commonly used clinical and pathologic features were found to have lower statistical significance and weaker or confounded effects -- tumor grade (P=0.026), number of nodes examined (P=0.040), and lymphovascular invasion (P=0.175).  

Consistency of the Colon Cancer Recurrence Score Across All Clinical Endpoints

The Recurrence Score was significantly associated with the primary clinical endpoint, recurrence-free interval (RFI) (P=0.004). In addition, the Recurrence Score was also significantly associated with both DFS (P=0.01) and OS (P=0.04). DFS and OS were defined as secondary clinical endpoints for the study because tumor biology would not be expected to predict death from causes other than colon cancer – events which are included in the DFS and OS endpoints.

Consistent and significant results for Recurrence Score across all three clinical endpoints studied (RFI, DFS, and OS) reinforce confidence in the association of the Recurrence Score with outcome in stage II colon cancer following surgery.

The Need for Individualized Risk Assessment: Limitations with Existing Clinical and Pathologic Markers

The colon cancer Recurrence Score result provides additional insight into the likelihood of colon cancer recurrence that goes beyond existing measures such as nodal assessment, bowel perforation, bowel obstruction, tumor grade, lymphatic or venous invasion, T-Stage and Mismatch Repair (MMR) Status. Current guidelines recommend that members with any of these clinical and pathologic risk features be considered high risk, with all other patients classified as standard risk. Lymph node involvement, T-stage, and MMR status have proven to be clear prognostic factors, yet apply to a minority of stage II colon cancer members.

T-Stage and Outcome in Stage II Colon Cancer – SEER database 1991-2000 

• T-Stage has been recognized as an important prognostic factor for stage II colon cancer.
• In 119,363 colon cancer patients from the SEER database, the 5-year risk of death was 15.3% in patients with T3 disease as compared to 27.8% for those with T4 tumors.
• Patients with T3 disease have a risk of death similar to the overall stage II population (15.3% versus 17% respectively).
• The unmet need remains with the 83% of stage II colon cancer patients with T3 tumors. These patients are known to be a heterogeneous population with a wide range of individual risk though as a group, they are currently considered to be at "standard risk". This proportion of T3 tumors in SEER is consistent with the 85% found in both the QUASAR study and multivariate QUASAR analysis.  

MMR/MSI and Outcome in Resected Colon Cancer 

• MMR Testing is helpful in identifying the 15% of stage II colon cancer members whose tumors are characterized by MMR deficiency (MMR-D). MMR testing may be performed by IHC for the MMR proteins (MLH1, MSH2) or by PCR for alterations in the lengths of DNA microsatellite sequences.
• MMR-D tumors are characterized by loss of expression of one or more MMR proteins by IHC as well as the presence of high-degree microsatellite instability by PCR (MSI-H). MMR-D is thus synonymous with MSI-H, and MMR proficiency (MMR-P) is synonymous with low-degree microsatellite instability (MSI-L) and microsatellite stability (MSS).
• In multiple studies, patients with resected colon cancer (particularly stage II) with MMR-D tumors have a significantly lower risk of recurrence compared to patients with MMR-P tumors.
• While there has been controversy as to whether MMR status predicts lack of benefit with 5FU-based adjuvant therapy, the evidence to date suggest that MMR-D stage II colon cancer members would derive little, if any, benefit from adjuvant chemotherapy, particularly given their very low baseline risk of recurrence.
• For the ~85% of members with MMR-P tumors, who would be considered standard risk, there is a need for clinical tools to more accurately determine which members should be appropriately considered for post-surgical chemotherapy for colon cancer.
• MMR testing remains controversial and there is no clear consensus on requiring this test as a prognostic marker.  

QUASAR Study: Insight Beyond Existing Markers 

1. The Recurrence Score result provides additional insight into the likelihood of colon cancer recurrence for members that go beyond existing measures. In the QUASAR validation study, an analysis was performed in 650 patients with available data for Recurrence Score, T stage, MMR status, tumor grade, number of nodes examined, and lymphovascular invasion. In this pre-specified multivariate analysis, Recurrence Score (P=0.008) demonstrated a significant and independent ability to predict recurrence, even after simultaneously accounting for the prognostic effects of tumor grade, lymphovascular invasion, number of nodes examined, T-stage, and MMR status. Recurrence Score, T stage and MMR status, were the most important, independent predictors of recurrence.
2. Given the independent, significant importance of T4 stage, MMR-D, and Recurrence Score for predicting recurrence risk in stage II colon cancer following surgery, these results support a paradigm in which individualized risk assessment can be achieved with knowledge of these three factors. Both T4 stage and MMR-D have limited applicability in terms of member percentages for the subgroups defined by these markers. The Recurrence Score provides a personalized risk analysis for each and every stage II member.
3. For members with T3, MMR-P tumors, the Recurrence Score provides information to further delineate risk in this population, where existing markers may not be informative. The above figure illustrates the Recurrence Score by recurrence risk relationship, adjusted for 3 subgroups of patients defined by T stage and MMR status.

4. Colon Cancer Recurrence Score Benefits: