Turning the promise of genomics into the practice of medicine
Featured Articles
How are stage II colon cancer patients being assessed for their recurrence risk today?
There are a limited set of clinical and pathologic markers that are in current practice and recommended by guidelines: obstruction or perforation, T stage, the number of nodes examined, tumor grade, lymphatic vascular invasion, and margin status (NCCN Colon Cancer Guidelines v2.2011). Existing markers of risk in stage II colon cancer identify small groups of patients with higher risk (e.g. presentation with bowel obstruction or perforation, T4 stage, presence of LVI, high tumor grade, fewer than 12 nodes examined, positive surgical margins). The majority of patients do not have a marker that categorizes them as higher risk (ie, tumors that are T3, MMR-P), and have risk at or near the average of the overall stage II population. Overall, a patient with stage II colon cancer has an average 5-year risk of death of 17.5%. Patients with T4 tumors (observed in specimens from 17% of stage II colon cancer) are at a significantly higher risk with 27.8% risk, but T3 patients are essentially comparable to the overall average, with 15.3% risk (O’Connell J Natl Cancer Inst 2004). As a result, existing markers are informative for a small group of patients who are identified of high-risk, but not for the majority of patients. Moreover, since patients are bucketed into two groups (higher risk and not higher risk), there is no systematic way to further discriminate risk for an individual patient based on these existing markers. This problem is particularly acute for the standard risk patient, who is defined by the complete absence of “high risk” features – by definition, there are no markers to further discriminate risk in this patient population, which is known to be clinically quite heterogeneous.
Current practice is based on the expectation that chemotherapy produces a similar proportional benefit for all patients. The paradigm in stage II colon cancer today is to treat patients with higher risk based on their clinical and pathologic features with the expectation that these patients, at higher baseline risk, derive a larger absolute benefit from chemotherapy.
The Oncotype DX Colon Cancer Assay
The Oncotype DX Colon Cancer Assay is a clinically validated diagnostic assay based on an individual patient’s colon tumor expression of 12 genes, and quantifies the likelihood of recurrence in stage II colon cancer following surgery. The continuous Recurrence Score® (RS) result has the greatest clinical utility for T3, Mismatch Repair Proficient (MMR–P) patients, who constitute the majority of patients diagnosed with stage II colon cancer:
The Oncotype DX<sup>®</sup> Colon Cancer Assay
Poster Presentations at ASCO-GI 2011
MMR by IHC: Agreement and Reproducibility
Assay result variability during determination of mismatch repair deficiency status using immunohistochemistry: A transatlantic comparative study (Abstract #403).
Gordon Hutchins, MD et al.

1)  Independent determination of MMR status by immunohistochemistry (IHC) on colorectal cancer tissue microarray (TMA) material was associated with good to excellent inter-laboratory and intra-observer agreement; the latter demonstrated excellent assay reproducibility.
2)  These data validate the routine use of IHC to determine MMR status, particularly since whole tissue IHC is less vulnerable to sampling heterogeneity when compared to TMAs.
2011 ASCO-GI Poster #403 – Hutchins
Grade as a Prognostic Factor in Stage II Colon Cancer
Reproducibility of colon tumor grade and relationship to recurrence in the context of clinical, pathologic, and genomic tumor features in 504 patients with stage II colon cancer treated with surgery alone at the Cleveland Clinic. (Abstract #526)
Ian Lavery, MD et al.

1)  High tumor grade was not found to be a marker of higher recurrence risk in stage II colon cancer by either of two pathologists using their standard clinical practice methods.

  • Contrary to conventional expectations, but consistent with other reported studies, high grade was associated in some circumstances with a lower risk of recurrence in stage II colon cancer.

2)   Inter-pathologist agreement on colon tumor grade was overall modest in this study, and moderate after excluding mucinous cases, even with central expert review.
3)  For the stage II colon cancer patient, recurrence risk should be assessed using T stage, MMR, and RS, the three key predictors of recurrence risk in stage II colon cancer, as reported in the ASCO* 2009 QUASAR presentation.
2011 ASCO-GI Poster #526 – Lavery
An economic study assessing Oncotype DX in stage II colon cancer management
Use of a multigene prognostic assay for selection of adjuvant chemotherapy in patients with stage II colon cancer: Impact on quality-adjusted life expectancy and costs. (Abstract #491)
Neal Meropol, MD et al.

1)   Clinical use of the Oncotype DX RS to assess risk of recurrence in T3 stage II colon cancers with intact MMR may improve quality-adjusted life expectancy and be cost-saving from a societal perspective.
2)  Patient age and disutility associated with chemotherapy are important considerations in adjuvant treatment decisions.
2011 ASCO-GI Poster #491 – Meropol
* ASCO® is a registered trademark of the American Society of Clinical Oncology. ASCO does not endorse any product or therapy.
Poster Presentation at ESMO 2010
Considerations in the Development and Validation of Genomic Tests for Cancer Recurrence and Treatment Benefit
Kerr DJ, O’Connell MJ, Lavery IC, et al.
Objective and Methods
This poster outlines the development and validation of the Oncotype DX® Colon Cancer Assay Recurrence Score® (RS) and Treatment Score (TS), describing the step-wise procedures behind these processes.
Tumor samples from four large studies (NSABP C04 and C06 (surgery plus adjuvant fluorouracil/folinic acid) and NSABP C01/C02 and a Cleveland Clinic cohort (surgery alone)) were used in the development and gene selection studies for the assay. The QUASAR clinical validation study was prospectively-designed and analyzed the 12-gene RS and 11-gene TS for an association with recurrence-free interval and differential treatment benefit, respectively. Data from both stage II and III patients were used in the development of the RS and TS, and clinical validation was performed in a study focused on stage II colon cancer.
For both RS and TS, gene expression was normalized to the 5 reference genes yielding a 12-gene RS and an 11–gene TS.
Results

1)  48 of 375 genes studied in all development studies were significantly associated with recurrence-free interval, the primary clinical endpoint, in both the surgery alone and at least one surgery+FU/FA study.
2)  68 of 375 genes studied in all development studies had significant interaction between gene expression and FU/FA treatment when data across the four studies were pooled.
3)  The final 7 RS and 6 TS genes were selected by evaluating the strength of association with recurrence, involvement in biologic pathways functionally important in colon cancer, consistency of performance across studies and in stage II and III patients, analytical performance, and a dynamic range of expression.
4)  The RS was significantly associated with RFI both before and after controlling for clinical and pathologic variables; RS by treatment interaction was not significant.
5) The TS was not significant in the validation study; there was no trend in absolute benefit in pre-specified direction across benefit groups. Most TS genes had a similar relationship with recurrence risk in both patients treated with surgery alone and surgery+FU/FA.
Conclusions
This strategy, involving more than 3000 patient specimens, successfully validated a gene expression signature for colon cancer recurrence but not FU/FA benefit.
The Recurrence Score was validated as an independent predictor of recurrence risk for individual stage II colon cancer patients. Patients with high Recurrence Score disease should be considered more strongly for adjuvant therapy due to higher recurrence risk and thus larger absolute risk reduction with FU/FA.
Investigation of genes which predict FU/FA benefit in this study may have been limited in part by lower statistical power for identification of these genes as well as use of non-randomized comparisons in development studies.
2010 ESMO Poster – Kerr
Spotlight on Oncotype DX Colon Cancer Assay Genes
In each newsletter we will describe the function of one of the Oncotype DX genes and its correlation to colon cancer recurrence.
Stromal Response Group : Tumor associated stroma is becoming increasingly recognized as an important factor in tumorgenesis; it is not an innocent bystander simply providing mechanical support to the tumor, but instead supplies a rich assortment of growth, inflammatory, angiogenic and invasion factors to the tumor. This activated (or "reactive") stroma is similar to what is seen during wound healing. Dvorak first described cancers as wounds that do not heal [1], and it is now generally accepted that activated stroma represents a "wound healing response" that can promote tumor growth, cell migration, invasion and angiogenesis. In the same way that tissue regeneration during wound healing involves a complex relationship between the host and the microenvironment, tumorgenesis is also dependant on extra-cellular interactions and signals from the stroma. TGF-beta signaling mediates fibroblast activation during wound healing and is therefore likely to play an important role in tumor-stroma communication during tumorgenesis.
Fibroblast Activation Protein (FAP): The FAP gene is one of the three genes in the stromal response group. FAP encodes the "Fibroblast Activation Protein", a cell surface bound enzyme which functions to cut peptide links in proteins (otherwise known as an integral membrane gelatinase, belonging to the serine protease family). FAP expression is detected in the stroma of over 90% of malignant breast, colorectal, lung, skin and pancreatic tumors, fibroblasts of healing wounds, soft tissue sarcomas, and some fetal mesenchymal cells, but shows limited or no expression in adult healthy tissue [2-6]. This membrane bound protease likely plays an important role in the complex signaling between tumor and the micro-environment which leads to invasion and metastasis. Colon cancers with high levels of FAP have more aggressive disease and a higher risk for metastasis and recurrence [7-8]. There is also evidence for this in pancreatic adenocarcinoma [9].
Since FAP is so highly expressed on activated tumor associated fibroblasts, but shows little expression in normal adult tissues, antibodies against this protein may have possible therapeutic applications [1, 10-11].

1)  Dvorak HF. Tumors: wounds that do not heal. Similarities between tumor stroma generation and wound healing.
2)  Garin-Chesa P, Old LJ, Rettig WJ. Cell surface glycoprotein of reactive stromal fibroblasts as a potential antibody target in human epithelial cancers. Proc Natl Acad Sci U S A 1990;87(18):7235–9.
3)  Scanlan MJ, Raj BK, Calvo B et al. Molecular cloning of fibroblast activation protein alpha, a member of the serine protease family selectively expressed in stromal fibroblasts of epithelial cancers. Proc Natl Acad Sci U S A 1994;91(12):5657–61.
4)  RRettig WJ, Garin-Chesa P, Beresford HR et al. Cell-surface glycoproteins of human sarcomas: differential expression in normal and malignant tissues and cultured cells. Proc Natl Acad Sci U S A 1988;85(9):3110–4.
5)  Bauer S, Jendro M, Wadle A, et al. Fibroblast activation protein is expressed by rheumatoid myofibroblast-like synoviocytes. Arthritis Res Ther 2006;8:R171.
6)  Park JE, Lenter MC, Zimmermann RN et al. Fibroblast activation protein, a dual specificity serine protease expressed in reactive human tumor stromal fibroblasts. J Biol Chem 1999;274:36505–36512.
N Engl J Med 1986, 315:1650-1659.
7)  Henry L, Lee H, Lee J, et al. Clinical implications of fibroblast activation protein in patients with colon cancer. Clin Cancer Res 2007, 13: 1736-41.
8)  O'Connell MJ, Lavery I, Yothers G, et al. Relationship between tumor gene expression and recurrence in four independent studies of stage II/III colon cancer patients treated with surgery alone and surgery plus adjuvant fluorouracil plus leucovorin. J Clin Oncol. 2010, 28(25):3904-7.
9)  Cohen SJ, Alpaugh RK, Palazzo I, et al. Fibroblast activation protein and its relationship to clinical outcome in pancreatic adenocarcinoma. Pancreas 2008, 37(2):154-8.
10)  Welt Divgi CR, Scott AM, Garin-Chesa P et al. Antibody targeting in metastatic colon cancer: a phase I study of monoclonal F19 against a cell-surface protein of reactive tumor stromal fibroblasts. J Clin Oncol. 1994, 12(6): 1193-203.
11)  Wolf BB, Quan C, Tran T, et al. On the edge of validation- cancer protease fibroblast activation protein. Mini Rev Med Chem. 2008, 8(7):719-27.
Genomic Health Turning the promise of genomics into the practice of medicine