- Email: [email protected]
Oncotype DX® colon cancer assay for prediction of recurrence risk in patients with stage II and III colon cancer: A review of the evidence
Surgical Oncology 24 (2015) 61e66
Contents lists available at ScienceDirect
Surgical Oncology journal homepage: www.elsevier.com/locate/suronc
Review
Oncotype DX® colon cancer assay for prediction of recurrence risk in patients with stage II and III colon cancer: A review of the evidence Y. Nancy You a, *, Rudolph B. Rustin b, James D. Sullivan c a
Department of Surgical Oncology, UT MD Anderson Cancer Center, Houston, TX, USA Mt. Pleasant, SC, USA c Hofstra North Shore-LIJ Medical School, Department of Surgery, Division of Surgical Oncology, Great Neck, NY, USA b
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 6 February 2015
Advances in molecular biology have enabled identification of tumor biomarkers that allow for individualized risk assessment for patients with cancer. Molecular predictors of clinical outcome can help inform discussion regarding the role of adjuvant chemotherapy in patients with resected colon cancer, such as those with stage II colon cancer in which the benefit of adjuvant therapy is controversial or those with stage III colon cancer who may have a lower risk of recurrence and less absolute benefit from oxaliplatin therapy. This article summarizes the data surrounding the development, validation, and clinical and economic utility of the Oncotype DX® colon cancer assay, a multigene expression assay validated to independently predict recurrence risk in patients with stage II and III colon cancer beyond traditional factors. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Colon cancer Oncotype DX 12-Gene assay Genomic assay Adjuvant chemotherapy Multigene assay Colon cancer recurrence risk Colon cancer treatment Oxaliplatin
Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Existing considerations in adjuvant chemotherapy for colon cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Development of the 12-gene colon cancer assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Validation of the 12-gene colon cancer assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 QUASAR (QUick and simple and reliable) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 CALGB 9581 (cancer and leukemia group B) study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 NSABP C-07 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Impact of the 12-gene colon cancer assay on clinical decision making . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Economic impact of the 12-gene colon cancer assay in clinical care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Discussion/conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Conflict of interest statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Authorship statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Introduction
* Corresponding author. Department of Surgical Oncology, UT MD Anderson Cancer Center, 1400 Hermann Pressler Drive, Unit 1484, Houston, TX 77030, USA. Tel.: þ1 713 794 4206; fax: þ1 713 792 4689. http://dx.doi.org/10.1016/j.suronc.2015.02.001 0960-7404/© 2015 Elsevier Ltd. All rights reserved.
Colorectal cancer remains the third leading cancer and the third leading cause of cancer death in both men and women in the United States, despite advances in treatment. There were over 130,000 new cases and over 50,000 deaths from colorectal cancer in 2014 alone [1]. The mainstay of treatment for non-metastatic
62
Y.N. You et al. / Surgical Oncology 24 (2015) 61e66
colon cancer is surgical resection of the tumor with adequate lymphadenectomy [2]. While surgery alone can be curative in 70% of patients with localized (stage II & III) colon cancer, a proportion of patients eventually develops tumor recurrence and ultimately dies of metastatic disease. Since surgeons are the main physicians of contact after tumor resection, we are challenged to answer questions regarding prognosis and need for further multi-modality treatments. Surgeons are furthermore responsible for making the decision to refer patients to medical oncologists for consideration of adjuvant chemotherapy. However, the benefit of adjuvant chemotherapy in resected stage II colon cancer remains controversial, and identifying which patients might derive the greatest benefit from adjuvant therapy for resected stage III disease is difficult. Biomarkers that can accurately assess the disease recurrence risk of an individual patient and better discriminate the absolute treatment benefits are needed. Specifically, it would be especially valuable to reliably identify stage II patients who are at higher risk of recurrence to have a greater absolute benefit with the addition of adjuvant chemotherapy; and it would be useful to stage III patients who are at lower risk for recurrence and may have less absolute benefit from oxaliplatin therapy to be spared its associated toxicity. Indeed, with advances in genomic medicine, several multigene assessment platforms are available today to help provide such information. The Oncotype DX® colon cancer assay, a 12-gene colon cancer assay, has been commercially available worldwide since 2010 and is the most established colon platforms available [12,13]. It was developed similarly to the multigene assay, Oncotype DX® Breast Cancer Assay (Genomic Health, Inc., Redwood City, CA), that has become part of the standard of care in breast cancer management, and has a wellestablished role in management guidelines published by the NCCN, St. Gallen Consensus, NICE, and the American Society of Clinical Oncology [8e11]. The purpose of this review article is to summarizing the development of and the clinical data validating this assay for the quantitative assessment of recurrence risk in the setting of colon cancer adjuvant-treatment. We also highlight potential uses of this assay in surgical practice for coordination of post-surgical recommendations, as well as the economic implications of its use. Existing considerations in adjuvant chemotherapy for colon cancer In patients with stage II colon cancer, adjuvant chemotherapy is currently being offered to patients considered to be at higher risk for disease relapse based on clinicopathologic factors, as they are presumed to derive greater benefit from adjuvant therapy. The National Comprehensive Cancer Network (NCCN) guidelines include a wide range of acceptable management strategies for stage II colon cancer: observation, single-agent fluoropyrimidine chemotherapy (i.e., 5-fluorouracil with leucovorin (5-FU/LV) or capecitabine), combination chemotherapy with oxaliplatin, and enrollment in clinical trials [2]. The guidelines recommend that treatment decisions about adjuvant therapy consider treatment benefits and risks, patient preferences, clinicopathologic features of high risk of systemic recurrence, and mismatch repair (MMR) status. High risk features have conventionally included tumor-stage 4 (T4) disease, low number of lymph nodes removed, bowel obstruction or perforation, lymphovascular invasion (LVI), positive or close margin status, high tumor grade, and perineural invasion [2]. In contrast, patients with tumors that are MMR-deficient (MMR-D) may have a good prognosis and do not appear to benefit from 5-FU chemotherapy [3]. Among these prognostic features, identification of T4 stage and MMR-D status are most helpful in defining risk for patients with stage II colon cancer. Each of these factors helps identify
approximately 15% of stage II patients at either end of the spectrum of recurrence risks. In the remaining approximately 70% of T3, MMR-proficient (MMR-P) patients with an average risk of recurrence, tumor grade, LVI and perineural invasion fail collectively to provide quantitation of recurrence risk. Furthermore, the determination of these histopathologic risk factors can be subjective and incomplete [2,4], while their combined use in a treatment algorithm has not been prospectively validated. These limitations of traditional histopathologic markers of risk underscore the need for new tools that allow surgeons to more accurately differentiate high-from low-risk patients with stage II colon cancer following tumor resection. For patients with stage III colon cancer, surgical resection followed by combination chemotherapy with 5-FU/LV plus oxaliplatin is the current standard recommendation [2]. Adding oxaliplatin to conventional fluoropyrimidine chemotherapy results in an approximate 20% relative risk reduction for disease-free survival (DFS), amounting to a 6% absolute benefit after 5 years. In addition, an exploratory analysis from the MOSAIC trial showed a trend toward improved 5-year DFS in a subset of stage II patients with highrisk features who received oxaliplatin in addition to 5-FU/LV [5]. However, oxaliplatin is associated with significant toxicity, including a risk of long-term peripheral neuropathy [6,7]. Since the absolute benefit from added oxaliplatin-based adjuvant chemotherapy is uncertain in stage II disease and may vary considerably in stage III disease, but oxaliplatin carries a significant risk of toxicities, a newer method for defining the risk of disease relapse may help the medical oncologist chose 5-FU vs. oxaliplatin-based regimens. Availability of molecular tools for risk stratification, together with intra-operative findings and clinicopathologic features, can help both the surgeon and the medical oncology team formulate an optimal and individualized approach to postsurgical care. Methods To identify literature utilized in this review, PubMed indexed literature was searched using key MeSH (Medical Subject Headings) Terms and Subheadings from the National Library of Medicine controlled vocabulary thesaurus used for indexing articles for PubMed. Only publications in English were considered. Searches were refined using the Advanced Search Builder. MeSH Terms/ Subheadings used included, but were not limited to the following: Oncotype DX Colon, Colonic Neoplasms/drug therapy; Colonic Neoplasms/genetics; Colonic Neoplasms/mortality; Colonic Neoplasms/pathology; Gene Expression Regulation, Neoplastic; Neoplasm Recurrence, Local; Neoplasm Recurrence, Local/prevention & control; Predictive Value of Tests; Prognosis; Tumor Markers, Biological/analysis; Tumor Markers, Biological/genetics; Chemotherapy, Adjuvant; Chemotherapy, Adjuvant/economics; CostBenefit Analysis; Quality-Adjusted Life Years. The authors reviewed the identified manuscripts for relevance to the study. Results A total of 12 studies were identified that described the development, validation, and clinical and economic utility of the 12-gene assay (Table 1). Development of the 12-gene colon cancer assay The 12-gene colon cancer assay for estimating risk of recurrence was developed through quantitative reverse transcription-PCR on RNA extracted from fixed, paraffin-embedded tumor blocks from 4 large, independent cohorts of patients with stage II or III colon
Y.N. You et al. / Surgical Oncology 24 (2015) 61e66
63
Table 1 Evidence supporting the Oncotype DX® colon cancer assay. Citation
N patients
Summary
Type of evidence
Clark-Langone, J Clin Oncol 2007
354
Development
O'Connell, J Clin Oncol 2010 Clark-Langone, BMC Cancer 2010 Gray, J Clin Oncol 2011
1851 N/A 1436
Venook, J Clin Oncol 2013
690
Yothers, J Clin Oncol 2013
892
Cartwright, Curr Med Res Opin 2013
92
Srivastava, The Oncologist 2014
141
Brenner, European Breast Cancer Conference 2013a Alberts, ASCO Gastrointestinal Cancers Symposium 2014a
269
Hornberger, Value Health 2012
1436
Alberts, Pharmacoeconomics 2014
141
Gene expression platform for biomarker discovery; 761 biomarkers screened Development of 12-gene algorithm Analytic and technical validation of assay Assay result significantly associated with recurrence risk beyond known prognostic factors for stage II patients Assay result significantly associated with recurrence risk beyond known prognostic factors for stage II patients Assay result significantly associated with recurrence risk beyond known prognostic factors for stage II and III patients Assay led to 29.3% overall change in treatment recommendations, in favor of less chemotherapy Assay led to 44.7% overall change in treatment recommendations, in favor of less chemotherapy Assay led to 37.9% overall change in treatment recommendations, in favor of less chemotherapy Assay influenced treatment decisions for a majority of patients and physicians, increased physician confidence, improved concordance in treatment choice between patients and physicians, and decreased patient decisional conflict Aassay was cost-saving without decreasing quality-adjusted life expectancy for patients with stage II, T3, MMR-P colon cancer Assay increased quality adjusted-survival and was cost-saving for patients with stage II, T3, MMR-P colon cancer
a
139
Development Analytic validation Clinical validation Clinical validation Clinical validation Clinical utility Clinical utility Clinical utility Clinical utility
Economic utility Economic utility
Newly-emerging data presented at scientific conferences.
cancer treated with surgery alone (National Surgical Adjuvant Breast and Bowel Project [NSABP] C-01/C-02, n ¼ 270; Cleveland Clinic, n ¼ 765) or surgery plus 5-FU and leucovorin (LV) (NSABP C04, n ¼ 308; NSABP C-06, n ¼ 508). A total of 761 genes were analyzed, and 48 were found to be significantly associated with risk of recurrence across multiple studies [14]. From these 48 genes, 7 cancer-related genes were selected on the basis of strength of association with patient outcomes in development studies, representation of key biologic pathways, consistency of performance across studies and patient subsets, and analytical performance. An algorithm was developed to integrate the expression levels of these 7 recurrence genes, normalized against 5 reference genes, into a 12gene assay yielding a Recurrence Score® value that predicts the risk of recurrence. The Recurrence Score result is a continuous variable and ranges from 0 to 100 (Fig. 1) [15]. An analytical validation study assessed the performance of the individual gene assays and the 12gene assay against pre-specified performance criteria to demonstrate that the assay was well-controlled and able to demonstrate consistent translation of specific tumor biology into clinically useful information [12].
Figure 1. The 12-gene colon cancer Recurrence Score and algorithm. The Recurrence Score gene panel consists of 12 genes: 7 cancer-related genes (FAP, INHBA, BGN, Ki-67, C-MYC, MYBL2, GADD45B) and 5 reference genes (ATP5E, GPX1, PGK1, UBB, VDAC2). The Recurrence Score is calculated using reference-normalized expression measurements and is scaled from 0 to 100.
Validation of the 12-gene colon cancer assay With the analytic procedures standardized, the 12-gene colon cancer assay was prospectively validated in 3 independent studies encompassing more than 3000 patients: QUASAR (stage II colon cancer, n ¼ 1436) [16]; CALGB 9581 (stage II colon cancer, n ¼ 690) [17]; NSABP C-07 (stage II/III colon cancer, 5-FU vs. 5-FU plus oxaliplatin, n ¼ 892) [18]. QUASAR (QUick and simple and reliable) QUASAR was a large (N ¼ 3239) randomized trial conducted between 1994 and 2003 to determine the survival benefit from adjuvant chemotherapy with 5-FU and LV vs. surgery alone in colon and rectal cancer [19]. From this trial, archival fixed, paraffinembedded, primary colon tumor blocks from 1436 patients with stage II colon cancer (surgery alone, n ¼ 711; surgery plus 5-FU/LV, n ¼ 725) were used in the prospectively designed validation study. The mean age of the population was 62 years (range, 30e83 years); 42% were women. Baseline patient and tumor characteristics were well balanced between the two treatment arms. For each tissue sample, the following measures were assessed centrally through review of H&E slides and pathology reports by pathologists blinded to treatment allocation and patient outcome: tumor site, histology subtype, grade, and T stage; number and status of lymph node samples; and LVI. MMR status was assessed by immunohistochemistry for MLH1 and MSH2 proteins at a single CLIA-certified laboratory [16]. The 12-gene colon cancer assay was then performed, with Recurrence Score results calculated from the prespecified algorithm (Fig. 1). The study demonstrated a significant association between risk of recurrence and continuous Recurrence Score values (P ¼ 0.004) in patients with stage II colon cancer randomized to surgery alone, with higher scores associated with higher recurrence risk (Fig. 2). Recurrence Score results were also significantly associated with disease-free survival (DFS) and OS. It is important to note that the clinical utility of the assay resides in the continuous nature of the Recurrence Score value, which provides 3-year recurrence risk
64
Y.N. You et al. / Surgical Oncology 24 (2015) 61e66
Figure 2. Colon cancer Recurrence Score result predicts recurrence following surgery. The graph shows Cox model estimates of recurrence risk at 3 years (with 95% confidence intervals [CIs]) by continuous Recurrence Score. There was a significant association between risk of recurrence and continuous Recurrence Score values (mean, 33.1; standard deviation [SD], 12.7; interquartile range [IQR], 18.2) in stage II colon cancer patients randomized to surgery alone, with a higher score associated with higher recurrence risk (hazard ratio [HR] per IQR, 1.38; 95% CI, 1.11e1.74; P ¼ 0.004). A rug plot depicting the distribution of recurrence score is included at the bottom of the graph.
estimates over a range from a low of approximately 9% to a high exceeding 25% [16]. In a multivariable analysis controlling for the effects of tumor grade, number of nodes examined, and LVI, the Recurrence Score, MMR status, and T stage were found to be the most significant, independent predictors of recurrence risk following surgery. Similar to previous reports, T4 tumors were associated with higher recurrence risk (HR 1.83, P ¼ 0.005) and MMR-D tumors were associated with a lower risk of recurrence (HR 0.32, P < 0.001) (Table 2) [16]. The Recurrence Score result predicted risk of recurrence beyond conventional markers with the greatest clinical utility in standard-risk patients with T3, MMR-P, stage II colon cancer (~70% of patients). These results support the value of a new, individualized approach to the assessment of risk of recurrence in stage II colon cancer on the basis of the Recurrence Score, MMR status, and T stage. CALGB 9581 (cancer and leukemia group B) study The CALGB 9581 parent study was a randomized trial comparing adjuvant treatment with edrecolomab, an experimental immunotherapeutic agent, against observation only in patients (n ¼ 1713) with resected, stage II colon cancer [20]. Edrecolomab showed no evidence of clinical efficacy. Compared with QUASAR, this trial had more patients who were 70 years of age and older (36%), whose Table 2 Analysis of association of multiple variables with risk of recurrence in patients with stage II colon cancer in QUASAR study (N ¼ 605). Variable
Categories
HR
95% CI
P-value
Recurrence score Mismatch repair T stage Tumor grade Nodes examined Lymphovascular Invasion
Continuous per 25 units Deficient vs proficient T4 vs T3 High vs low <12 vs 12 Present vs absent
1.43 0.36 1.87 0.65 1.43 1.43
(1.11, (0.17, (1.25, (0.42, (0.98, (0.90,
0.006 0.004 0.004 0.050 0.058 0.150
1.83) 0.80) 2.81) 1.02) 2.08) 2.28)
Abbreviations: CI, confidence interval; LVI, lymphovascular invasion; MMR, mismatch repair. Adapted from Gray RG, Quirke P, Handley K et al. J Clin Oncol. 2011; 29(35):4611e9.
tumors were MMR-D (21%), and who had, on average, lower risk of recurrence (approximately 15% at 5 years). For these reasons, archived tumor specimens from this trial provided a unique opportunity to validate the 12-gene colon cancer assay in a relatively low-risk population. Tissue was available for 1137 (68%) of 1672 stage II colon cancer patients from the CALGB 9581 study. The final study population included 690 patients from both study arms, including all available patients with recurrences (162 events) and a random sample of patients without recurrence (1:3 ratio). The Recurrence Score was calculated using the same prespecified gene list and algorithm previously tested in the QUASAR validation study (Fig. 1) [17]. Consistent with QUASAR, in the primary analysis there was a significant association between the Recurrence Score result and recurrence risk (P ¼ 0.013) in patients with stage II colon cancer. A pre-specified multivariate analysis showed that the Recurrence Score result independently predicted recurrence risk (P ¼ 0.004) after controlling for MMR status, T stage, number of nodes examined, histologic grade, and LVI. In a univariate analysis MMR status was the only covariate significantly associated with recurrence free interval (P ¼ 0.04). Even in this relatively low-risk population, the 12-gene assay was able to identify 22% of patients with T3 MMR-P tumors with an average 5-year risk of recurrence of 21% [17]. These results provide further evidence of the ability of the Recurrence Score result to quantify the risk of recurrence in stage II colon cancer and to further discriminate risk level in patients with T3, MMR-P tumors, which may help to identify patients for whom adjuvant chemotherapy may be considered. NSABP C-07 More recently, the performance of the 12-gene colon cancer assay was validated in a prospectively designed study using archived tissue (n ¼ 892) and prespecified endpoints, methods, and analysis from patients with stages II and III colon cancer randomized to 5-FU or 5-FU plus oxaliplatin in the NSABP C-07 study. NSABP C-07 was a randomized phase III clinical trial that evaluated the efficacy and safety of adding oxaliplatin to bolus 5-FU/LV chemotherapy following surgical resection in patients with stage II and III colon cancer [21]. Among the 892 patients evaluated, a total of 264 (29.6%) patients had stage II disease, 409 (45.9%) patients had stage IIIA/B disease, and 219 (24.6%) had stage IIIC disease. In this study, the continuous Recurrence Score result was significantly associated with a risk of recurrence after controlling for the effects of treatment and stage (P < 0.001). The Recurrence Score assay performed similarly in stages II and III colon cancer and was significantly associated with risk of recurrence after adjusting for T and N stage, number of nodes examined, MMR status, histologic grade, and treatment (P < 0.001) [18]. With the relative benefit of oxaliplatin being similar across the range of Recurrence Score values, the absolute benefit of oxaliplatin therefore increased with higher Recurrence Score results. The increase in absolute benefit between patients with high and low Recurrence Score results was most pronounced in stage II and IIIA/B disease. These results extend the findings of the QUASAR and CALGB 9581 studies and further validate the performance of the 12-gene colon cancer assay in stage II and stage III colon cancer patients treated with 5-FU and 5FU þ oxalipaltin. Impact of the 12-gene colon cancer assay on clinical decision making While oxaliplatin/5-FU-based regimens may improve DFS and OS, the toxic effects of such regimens, such as peripheral neuropathy, have been shown to persist as long as 4 years after
Y.N. You et al. / Surgical Oncology 24 (2015) 61e66
treatment [5]. The utility of the Recurrence Score result in clinical practice allows high-risk and low-risk patients to be distinguished so that high-risk patients can receive therapies that offer a larger potential absolute benefit, whereas low-risk patients can be spared unnecessary treatment that may offer only minimal benefit. The Recurrence Score result enables discrimination of absolute benefit as a function of risk. Accordingly, surgeons and medical oncologists can inform their patients with lower Recurrence Score results about the trade-offs between smaller absolute benefits and the risks of chemotherapy-related toxicities, while patients with higher scores may be candidates for chemotherapy treatment as they would be expected to derive greater absolute benefit. The impact of the 12-gene colon cancer assay results on stage II colon cancer in clinical practice was evaluated after the first two years of use in a survey of medical oncologists [22]. Subsequently, two prospective studies examined the impact of the assay on clinical decision making at multiple centers within the Mayo Clinic Cancer Research Consortium network [23], and in patients from the Clalit Health Services in Israel [24]. Three hundred forty-six U.S. medical oncologists who had ordered the 12-gene assay completed a 34-item survey from December 2010 to December 2011. Among 92 physicians who reported pre- and post-assay recommendations based upon chart reviews, 27 (29%) changed their recommendations after receiving the 12-gene colon cancer assay results: 2/3 of the treatment recommendations decreased in intensity, whereas 1/3 increased in intensity. These changes were in agreement with Recurrence Score results, in that treatment intensity decreased more often for lower Recurrence Score values and increased for higher scores (P ¼ 0.0035 for the trend). These results suggest that use of the assay in clinical practice may be associated with meaningful changes in treatment recommendations for patients with stage II colon cancerdspecifically, reductions in treatment intensity that could spare patients unnecessary toxic effects of chemotherapy [22]. These results were later confirmed in the first prospective study to assess the role of the quantitative Recurrence Score result in clinical decision-making for patients with stage IIA, MMR-P colon cancer. This multicenter prospective study was conducted by more than 100 oncologists within the Mayo Clinic Cancer Research Consortium network who enrolled patients with stage IIA (T3N0) colon cancer across 17 academic and community sites in the study to characterize the impact of Recurrence Score results on medical oncologists' recommendations regarding adjuvant chemotherapy in T3, MMR-P, stage II colon cancer patients. Tumor specimens from patients with resected stage IIA colon cancer who were candidates for adjuvant chemotherapy were assessed using the Recurrence Score value and MMR status. Physicians completed surveys prior to and after receiving the Recurrence Score report along with MMR results and recorded their planned treatments. In the primary analysis treatment recommendations changed for 63 (45%) of 141 MMR-P patients with most (3/4) changes in the direction of decreased intensity. Recommendations for chemotherapy decreased from 52% of patients pre-assay to 30% postassay. The direction of change was concordant with the Recurrence Score result (p ¼ 0.011). These findings indicate that use of the 12-gene assay may lead to overall reductions in the use of adjuvant chemotherapy in T3, MMR-P, stage II colon cancer patients [23]. Secondary analyses were carried out to evaluate patient treatment decisions, physician confidence, concordance in treatment choice between physicians and patients, and patient decisional conflict after receiving the 12-gene assay result. One hundred ninety of 221 patients enrolled were evaluable, including 139 with MMR-P tumors. Post-assay treatment decisions in 129 (96%) of 135
65
patients were concordant between patients and physicians compared to 49 (66%) of 74 treatment decisions prior to receiving the 12-gene assay result. In the majority of cases, patients (85%) and physicians (69%) reported that the assay influenced their treatment decisions. Further, the 12-gene assay results increased physician confidence in their treatment recommendations in 126 (84%) and provided additional clinically relevant information to physicians in 129 (86%) of cases. Patient decisional conflict was significantly lower after assay results (p < 0.001) [25]. In the Clalit Health Services study, investigators prospectively collected treatment recommendations before testing with the 12gene assay and actual treatment received following receipt of test results in T3 MMR-P stage II colon cancer patients between January 2011 and May 2012. Pre-assay treatment recommendations were compared with assay results and actual treatment delivered. Preassay treatment recommendations differed from post-assay treatment delivered in 102 (38%) of 269 evaluable patients, with less treatment delivered for 76 (28%) and more for 26 (10%). Pre-assay, 121 (45%) patients received recommendation for chemotherapy; post-assay, 75 (28%) actually received chemotherapy. Changes in treatment intensity were generally consistent with the Recurrence Score results with increased treatment intensity observed more often at higher scores and decreased intensity at lower scores [24]. Consistent with prior results, in this first international study of the role of the 12-gene assay in clinical decision making, the use of the 12-gene assay had an impact on treatment decisions in Israel, with an overall net reduction in chemotherapy use. Economic impact of the 12-gene colon cancer assay in clinical care In 2012 an economic analyses comparing the 12-gene assay to published patterns of care from the NCCN® Colon/Rectum Cancer Outcomes Database predicted that the assay would improve outcomes while lowering medical costs for stage II, T3, MMR proficient patients [26]. Health economic analysis from this study projected that use of the Recurrence Score result to guide treatment decisions could reduce the use of adjuvant chemotherapy by an average of 17%, result in an average decrease in direct medical costs of nearly $3000 per patient, and increase quality-adjusted life expectancy by an average of 0.035 years [26]. A more recent study [26] assessed the validity of these findings using data from the prospective Mayo Clinic Cancer Research Consortium clinical utility study discussed above. Results of this study revealed a decrease in recommendations towards any adjuvant chemotherapy use by 22%, with a decrease in recommendations towards 5-FU þ oxaliplatin adjuvant chemotherapy use of 11%. Quality-adjusted survival increased on average by 0.114 years, and average lifetime medical savings was estimated at $991 per patient [27]. The cost savings to patients was attributed to an anticipated lower total cost of adjuvant chemotherapy resulting from decreases in cost associated with drugs, administration and adverse events management. These savings were expected to persist even if the cost of oxaliplatin dropped by 75% as a result of generic substitution. These analyses confirm the results of the earlier economic analysis and show that use of the 12-gene assay increases quality-adjusted survival and may be economically favorable for patients with stage II, T3, MMR-P colon cancer. Discussion/conclusion Because the role of adjuvant therapy in patients with stage II colon cancer remains controversial, yet the toxicity risk of chemotherapy is well-known and real, the optimal treatment plan should be tailored to each patient's level of individual risk. Surgeons, who provide the mainstay of treatment for patients with stage II and III colon cancer, namely, oncologic resection of tumor, may be involved
66
Y.N. You et al. / Surgical Oncology 24 (2015) 61e66
in multidisciplinary tumor boards with medical oncologists to make recommendations for adjuvant therapy and/or post-resection surveillance. Therefore, there is a critical need for tools that allow surgeons to accurately assess each patient's risk of recurrence and expected absolute benefit from adjuvant chemotherapy. Pathologic tumor stage and MMR status, along with other pathologic features, are conventionally used as the first steps in the risk assessment; but molecular-based diagnostic tests could be used to reflect the underlying tumor biology and provide more precise estimates of recurrence risk. An area of great practical benefit from such as individualized risk assessment assay includes standard-risk patients with T3, MMR-P, stage II colon cancer, for whom conventional prognostic factors are not informative. The 12-gene colon cancer assay in stage II colon cancer exemplifies a tool for personalized risk assessment that can aid the surgeon and medical oncologist in assessing recurrence risk, and deciding about adjuvant chemotherapy, while optimizing the quality-adjusted survival. Another area of clinical benefit is the discrimination of absolute oxaliplatin benefit as a function of risk in patients with stage III colon cancer [18]. Patients with lower Recurrence Score results, particularly those of older age, competing comorbidities, and/or concerns about quality of life, may be counseled about the trade-offs between smaller absolute benefits and the risks of oxaliplatin-related toxicities. In summary, the 12-gene colon cancer assay provides clinical utility to surgeons and medical oncologists making decisions about oxaliplatin and/or 5-FU based adjuvant chemotherapy in patients with stage II and also stage III colon cancer. Conflict of interest statement Y. You, R. Rustin, and J. Sullivan received an honorarium for their time and efforts. Authorship statement Guarantor of the integrity of the study: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Study concepts: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Study design: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Definition of intellectual content: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Literature research: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Clinical studies: Y. Nancy You Data acquisition: Y. Nancy You Data analysis: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Manuscript preparation: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Manuscript editing: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Manuscript review: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin References [1] Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics. CA Cancer J Clin 2014;2014(64):9e29. [2] NCCN. NCCN clinical practice guidelines in oncology: colon cancer version 2. 2015. ed2015. [3] Sargent DJ, Marsoni S, Monges G, Thibodeau SN, Labianca R, Hamilton SR, et al. Defective mismatch repair as a predictive marker for lack of efficacy of fluorouracil-based adjuvant therapy in colon cancer. J Clin Oncol 2010;28: 3219e26. [4] Midgley R, Rasul K, Al Salama H, Kerr DJ. Gene profiling in early stage disease. Cancer J 2010;16:210e3.
[5] Andre T, Boni C, Navarro M, Tabernero J, Hickish T, Topham C, et al. Improved overall survival with oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment in stage II or III colon cancer in the MOSAIC trial. J Clin Oncol 2009;27:3109e16. [6] Kuebler JP, Colangelo L, O'Connell MJ, Smith RE, Yothers G, Begovic M, et al. Severe enteropathy among patients with stage II/III colon cancer treated on a randomized trial of bolus 5-fluorouracil/leucovorin plus or minus oxaliplatin: a prospective analysis. Cancer 2007;110:1945e50. [7] Land SR, Kopec JA, Cecchini RS, Ganz PA, Wieand HS, Colangelo LH, et al. Neurotoxicity from oxaliplatin combined with weekly bolus fluorouracil and leucovorin as surgical adjuvant chemotherapy for stage II and III colon cancer: NSABP C-07. J Clin Oncol 2007;25:2205e11. [8] NCCN. NCCN clinical practice guidelines in oncology: breast cancer. v3. 2014. ed2014. [9] Harris L, Fritsche H, Mennel R, Norton L, Ravdin P, Taube S, et al. American society of clinical oncology 2007 update of recommendations for the use of tumor markers in breast cancer. J Clin Oncol 2007;25:5287e312. [10] Goldhirsch A, Winer EP, Coates AS, Gelber RD, Piccart-Gebhart M, Thurlimann B, et al. Personalizing the treatment of women with early breast cancer: highlights of the St Gallen International Expert Consensus on the primary therapy of early breast cancer 2013. Ann Oncol 2013;24:2206e23. [11] Gene expression profiling and expanded immunohistochemistry tests for guiding adjuvant chemotherapy decisions in early breast cancer management: MammaPrint, Oncotype DX, IHC4 and Mammostrat. National Institute for Health and Care Excellence; 2013. [12] Clark-Langone KM, Sangli C, Krishnakumar J, Watson D. Translating tumor biology into personalized treatment planning: analytical performance characteristics of the oncotype DX colon cancer assay. BMC Cancer 2010;10:691. [13] Webber EM, Lin JS, Evelyn PW. Oncotype DX tumor gene expression profiling in stage II colon cancer. Application: prognostic, risk prediction. PLoS Curr 2010:2. [14] Clark-Langone KM, Wu JY, Sangli C, Chen A, Snable JL, Nguyen A, et al. Biomarker discovery for colon cancer using a 761 gene RT-PCR assay. BMC Genomics 2007;8:279. [15] O'Connell MJ, Lavery I, Yothers G, Paik S, Clark-Langone KM, Lopatin M, et al. Relationship between tumor gene expression and recurrence in four independent studies of patients with stage II/III colon cancer treated with surgery alone or surgery plus adjuvant fluorouracil plus leucovorin. J Clin Oncol 2010;28:3937e44. [16] Gray RG, Quirke P, Handley K, Lopatin M, Magill L, Baehner FL, et al. Validation study of a quantitative multigene reverse transcriptase-polymerase chain reaction assay for assessment of recurrence risk in patients with stage II colon cancer. J Clin Oncol 2011;29:4611e9. [17] Venook AP, Niedzwiecki D, Lopatin M, Ye X, Lee M, Friedman PN, et al. Biologic determinants of tumor recurrence in stage II colon cancer: validation study of the 12-gene recurrence score in cancer and leukemia group B (CALGB) 9581. J Clin Oncol 2013;31:1775e81. [18] Yothers G, O'Connell MJ, Lee M, Lopatin M, Clark-Langone KM, Millward C, et al. Validation of the 12-gene colon cancer recurrence score in NSABP C-07 as a predictor of recurrence in patients with stage II and III colon cancer treated with fluorouracil and leucovorin (FU/LV) and FU/LV plus oxaliplatin. J Clin Oncol 2013;31:4512e9. [19] Quasar Collaborative G, Gray R, Barnwell J, McConkey C, Hills RK, Williams NS, et al. Adjuvant chemotherapy versus observation in patients with colorectal cancer: a randomised study. Lancet 2007;370:2020e9. [20] Niedzwiecki D, Bertagnolli MM, Warren RS, Compton CC, Kemeny NE, Benson 3rd AB, et al. Documenting the natural history of patients with resected stage II adenocarcinoma of the colon after random assignment to adjuvant treatment with edrecolomab or observation: results from CALGB 9581. J Clin Oncol 2011;29:3146e52. [21] Kuebler JP, Wieand HS, O'Connell MJ, Smith RE, Colangelo LH, Yothers G, et al. Oxaliplatin combined with weekly bolus fluorouracil and leucovorin as surgical adjuvant chemotherapy for stage II and III colon cancer: results from NSABP C-07. J Clin Oncol 2007;25:2198e204. [22] Cartwright T, Chao C, Lee M, Lopatin M, Bentley T, Broder M, et al. Effect of the 12-gene colon cancer assay results on adjuvant treatment recommendations in patients with stage II colon cancer. Curr Med Res Opin 2014;30:321e8. [23] Srivastava G, Renfro LA, Behrens RJ, Lopatin M, Chao C, Soori GS, et al. Prospective multicenter study of the impact of oncotype DX colon cancer assay results on treatment recommendations in stage II colon cancer patients. Oncologist 2014;19:492e7. [24] Brenner B, Lopatin M, Lee M, Geva R, Beny A, Dror Y, et al. Impact of the 12gene colon cancer recurrence score assay on clinical decision-making for adjuvant therapy in stage II colon cancer patients in Israel. In: European Cancer conference. Amsterdam, Netherlands; 2012. [25] Alberts SR, Renfro LA, Lopatin M, Tezcan H, Sloan J, Chao C, et al. Prospective evaluation of a 12-gene assay on patient treatment decisions and physician confidence in mismatch repair-proficient (MMR-P) stage IIa colon cancer patients. In: ASCO gastrointestinal cancers symposium. San Francisco, CA; 2014. [26] Hornberger J, Lyman GH, Chien R, Meropol NJ. A multigene prognostic assay for selection of adjuvant chemotherapy in patients with T3, stage II colon cancer: impact on quality-adjusted life expectancy and costs. Value Health 2012;15:1014e21. [27] Alberts SR, Yu TM, Behrens RJ, Renfro LA, Srivastava G, Soori GS, et al. Comparative economics of a 12-gene assay for predicting risk of recurrence in stage II colon cancer. Pharmacoeconomics 2014;32:1231e43.
Contents lists available at ScienceDirect
Surgical Oncology journal homepage: www.elsevier.com/locate/suronc
Review
Oncotype DX® colon cancer assay for prediction of recurrence risk in patients with stage II and III colon cancer: A review of the evidence Y. Nancy You a, *, Rudolph B. Rustin b, James D. Sullivan c a
Department of Surgical Oncology, UT MD Anderson Cancer Center, Houston, TX, USA Mt. Pleasant, SC, USA c Hofstra North Shore-LIJ Medical School, Department of Surgery, Division of Surgical Oncology, Great Neck, NY, USA b
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 6 February 2015
Advances in molecular biology have enabled identification of tumor biomarkers that allow for individualized risk assessment for patients with cancer. Molecular predictors of clinical outcome can help inform discussion regarding the role of adjuvant chemotherapy in patients with resected colon cancer, such as those with stage II colon cancer in which the benefit of adjuvant therapy is controversial or those with stage III colon cancer who may have a lower risk of recurrence and less absolute benefit from oxaliplatin therapy. This article summarizes the data surrounding the development, validation, and clinical and economic utility of the Oncotype DX® colon cancer assay, a multigene expression assay validated to independently predict recurrence risk in patients with stage II and III colon cancer beyond traditional factors. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Colon cancer Oncotype DX 12-Gene assay Genomic assay Adjuvant chemotherapy Multigene assay Colon cancer recurrence risk Colon cancer treatment Oxaliplatin
Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Existing considerations in adjuvant chemotherapy for colon cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Development of the 12-gene colon cancer assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Validation of the 12-gene colon cancer assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 QUASAR (QUick and simple and reliable) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 CALGB 9581 (cancer and leukemia group B) study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 NSABP C-07 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Impact of the 12-gene colon cancer assay on clinical decision making . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Economic impact of the 12-gene colon cancer assay in clinical care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Discussion/conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Conflict of interest statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Authorship statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Introduction
* Corresponding author. Department of Surgical Oncology, UT MD Anderson Cancer Center, 1400 Hermann Pressler Drive, Unit 1484, Houston, TX 77030, USA. Tel.: þ1 713 794 4206; fax: þ1 713 792 4689. http://dx.doi.org/10.1016/j.suronc.2015.02.001 0960-7404/© 2015 Elsevier Ltd. All rights reserved.
Colorectal cancer remains the third leading cancer and the third leading cause of cancer death in both men and women in the United States, despite advances in treatment. There were over 130,000 new cases and over 50,000 deaths from colorectal cancer in 2014 alone [1]. The mainstay of treatment for non-metastatic
62
Y.N. You et al. / Surgical Oncology 24 (2015) 61e66
colon cancer is surgical resection of the tumor with adequate lymphadenectomy [2]. While surgery alone can be curative in 70% of patients with localized (stage II & III) colon cancer, a proportion of patients eventually develops tumor recurrence and ultimately dies of metastatic disease. Since surgeons are the main physicians of contact after tumor resection, we are challenged to answer questions regarding prognosis and need for further multi-modality treatments. Surgeons are furthermore responsible for making the decision to refer patients to medical oncologists for consideration of adjuvant chemotherapy. However, the benefit of adjuvant chemotherapy in resected stage II colon cancer remains controversial, and identifying which patients might derive the greatest benefit from adjuvant therapy for resected stage III disease is difficult. Biomarkers that can accurately assess the disease recurrence risk of an individual patient and better discriminate the absolute treatment benefits are needed. Specifically, it would be especially valuable to reliably identify stage II patients who are at higher risk of recurrence to have a greater absolute benefit with the addition of adjuvant chemotherapy; and it would be useful to stage III patients who are at lower risk for recurrence and may have less absolute benefit from oxaliplatin therapy to be spared its associated toxicity. Indeed, with advances in genomic medicine, several multigene assessment platforms are available today to help provide such information. The Oncotype DX® colon cancer assay, a 12-gene colon cancer assay, has been commercially available worldwide since 2010 and is the most established colon platforms available [12,13]. It was developed similarly to the multigene assay, Oncotype DX® Breast Cancer Assay (Genomic Health, Inc., Redwood City, CA), that has become part of the standard of care in breast cancer management, and has a wellestablished role in management guidelines published by the NCCN, St. Gallen Consensus, NICE, and the American Society of Clinical Oncology [8e11]. The purpose of this review article is to summarizing the development of and the clinical data validating this assay for the quantitative assessment of recurrence risk in the setting of colon cancer adjuvant-treatment. We also highlight potential uses of this assay in surgical practice for coordination of post-surgical recommendations, as well as the economic implications of its use. Existing considerations in adjuvant chemotherapy for colon cancer In patients with stage II colon cancer, adjuvant chemotherapy is currently being offered to patients considered to be at higher risk for disease relapse based on clinicopathologic factors, as they are presumed to derive greater benefit from adjuvant therapy. The National Comprehensive Cancer Network (NCCN) guidelines include a wide range of acceptable management strategies for stage II colon cancer: observation, single-agent fluoropyrimidine chemotherapy (i.e., 5-fluorouracil with leucovorin (5-FU/LV) or capecitabine), combination chemotherapy with oxaliplatin, and enrollment in clinical trials [2]. The guidelines recommend that treatment decisions about adjuvant therapy consider treatment benefits and risks, patient preferences, clinicopathologic features of high risk of systemic recurrence, and mismatch repair (MMR) status. High risk features have conventionally included tumor-stage 4 (T4) disease, low number of lymph nodes removed, bowel obstruction or perforation, lymphovascular invasion (LVI), positive or close margin status, high tumor grade, and perineural invasion [2]. In contrast, patients with tumors that are MMR-deficient (MMR-D) may have a good prognosis and do not appear to benefit from 5-FU chemotherapy [3]. Among these prognostic features, identification of T4 stage and MMR-D status are most helpful in defining risk for patients with stage II colon cancer. Each of these factors helps identify
approximately 15% of stage II patients at either end of the spectrum of recurrence risks. In the remaining approximately 70% of T3, MMR-proficient (MMR-P) patients with an average risk of recurrence, tumor grade, LVI and perineural invasion fail collectively to provide quantitation of recurrence risk. Furthermore, the determination of these histopathologic risk factors can be subjective and incomplete [2,4], while their combined use in a treatment algorithm has not been prospectively validated. These limitations of traditional histopathologic markers of risk underscore the need for new tools that allow surgeons to more accurately differentiate high-from low-risk patients with stage II colon cancer following tumor resection. For patients with stage III colon cancer, surgical resection followed by combination chemotherapy with 5-FU/LV plus oxaliplatin is the current standard recommendation [2]. Adding oxaliplatin to conventional fluoropyrimidine chemotherapy results in an approximate 20% relative risk reduction for disease-free survival (DFS), amounting to a 6% absolute benefit after 5 years. In addition, an exploratory analysis from the MOSAIC trial showed a trend toward improved 5-year DFS in a subset of stage II patients with highrisk features who received oxaliplatin in addition to 5-FU/LV [5]. However, oxaliplatin is associated with significant toxicity, including a risk of long-term peripheral neuropathy [6,7]. Since the absolute benefit from added oxaliplatin-based adjuvant chemotherapy is uncertain in stage II disease and may vary considerably in stage III disease, but oxaliplatin carries a significant risk of toxicities, a newer method for defining the risk of disease relapse may help the medical oncologist chose 5-FU vs. oxaliplatin-based regimens. Availability of molecular tools for risk stratification, together with intra-operative findings and clinicopathologic features, can help both the surgeon and the medical oncology team formulate an optimal and individualized approach to postsurgical care. Methods To identify literature utilized in this review, PubMed indexed literature was searched using key MeSH (Medical Subject Headings) Terms and Subheadings from the National Library of Medicine controlled vocabulary thesaurus used for indexing articles for PubMed. Only publications in English were considered. Searches were refined using the Advanced Search Builder. MeSH Terms/ Subheadings used included, but were not limited to the following: Oncotype DX Colon, Colonic Neoplasms/drug therapy; Colonic Neoplasms/genetics; Colonic Neoplasms/mortality; Colonic Neoplasms/pathology; Gene Expression Regulation, Neoplastic; Neoplasm Recurrence, Local; Neoplasm Recurrence, Local/prevention & control; Predictive Value of Tests; Prognosis; Tumor Markers, Biological/analysis; Tumor Markers, Biological/genetics; Chemotherapy, Adjuvant; Chemotherapy, Adjuvant/economics; CostBenefit Analysis; Quality-Adjusted Life Years. The authors reviewed the identified manuscripts for relevance to the study. Results A total of 12 studies were identified that described the development, validation, and clinical and economic utility of the 12-gene assay (Table 1). Development of the 12-gene colon cancer assay The 12-gene colon cancer assay for estimating risk of recurrence was developed through quantitative reverse transcription-PCR on RNA extracted from fixed, paraffin-embedded tumor blocks from 4 large, independent cohorts of patients with stage II or III colon
Y.N. You et al. / Surgical Oncology 24 (2015) 61e66
63
Table 1 Evidence supporting the Oncotype DX® colon cancer assay. Citation
N patients
Summary
Type of evidence
Clark-Langone, J Clin Oncol 2007
354
Development
O'Connell, J Clin Oncol 2010 Clark-Langone, BMC Cancer 2010 Gray, J Clin Oncol 2011
1851 N/A 1436
Venook, J Clin Oncol 2013
690
Yothers, J Clin Oncol 2013
892
Cartwright, Curr Med Res Opin 2013
92
Srivastava, The Oncologist 2014
141
Brenner, European Breast Cancer Conference 2013a Alberts, ASCO Gastrointestinal Cancers Symposium 2014a
269
Hornberger, Value Health 2012
1436
Alberts, Pharmacoeconomics 2014
141
Gene expression platform for biomarker discovery; 761 biomarkers screened Development of 12-gene algorithm Analytic and technical validation of assay Assay result significantly associated with recurrence risk beyond known prognostic factors for stage II patients Assay result significantly associated with recurrence risk beyond known prognostic factors for stage II patients Assay result significantly associated with recurrence risk beyond known prognostic factors for stage II and III patients Assay led to 29.3% overall change in treatment recommendations, in favor of less chemotherapy Assay led to 44.7% overall change in treatment recommendations, in favor of less chemotherapy Assay led to 37.9% overall change in treatment recommendations, in favor of less chemotherapy Assay influenced treatment decisions for a majority of patients and physicians, increased physician confidence, improved concordance in treatment choice between patients and physicians, and decreased patient decisional conflict Aassay was cost-saving without decreasing quality-adjusted life expectancy for patients with stage II, T3, MMR-P colon cancer Assay increased quality adjusted-survival and was cost-saving for patients with stage II, T3, MMR-P colon cancer
a
139
Development Analytic validation Clinical validation Clinical validation Clinical validation Clinical utility Clinical utility Clinical utility Clinical utility
Economic utility Economic utility
Newly-emerging data presented at scientific conferences.
cancer treated with surgery alone (National Surgical Adjuvant Breast and Bowel Project [NSABP] C-01/C-02, n ¼ 270; Cleveland Clinic, n ¼ 765) or surgery plus 5-FU and leucovorin (LV) (NSABP C04, n ¼ 308; NSABP C-06, n ¼ 508). A total of 761 genes were analyzed, and 48 were found to be significantly associated with risk of recurrence across multiple studies [14]. From these 48 genes, 7 cancer-related genes were selected on the basis of strength of association with patient outcomes in development studies, representation of key biologic pathways, consistency of performance across studies and patient subsets, and analytical performance. An algorithm was developed to integrate the expression levels of these 7 recurrence genes, normalized against 5 reference genes, into a 12gene assay yielding a Recurrence Score® value that predicts the risk of recurrence. The Recurrence Score result is a continuous variable and ranges from 0 to 100 (Fig. 1) [15]. An analytical validation study assessed the performance of the individual gene assays and the 12gene assay against pre-specified performance criteria to demonstrate that the assay was well-controlled and able to demonstrate consistent translation of specific tumor biology into clinically useful information [12].
Figure 1. The 12-gene colon cancer Recurrence Score and algorithm. The Recurrence Score gene panel consists of 12 genes: 7 cancer-related genes (FAP, INHBA, BGN, Ki-67, C-MYC, MYBL2, GADD45B) and 5 reference genes (ATP5E, GPX1, PGK1, UBB, VDAC2). The Recurrence Score is calculated using reference-normalized expression measurements and is scaled from 0 to 100.
Validation of the 12-gene colon cancer assay With the analytic procedures standardized, the 12-gene colon cancer assay was prospectively validated in 3 independent studies encompassing more than 3000 patients: QUASAR (stage II colon cancer, n ¼ 1436) [16]; CALGB 9581 (stage II colon cancer, n ¼ 690) [17]; NSABP C-07 (stage II/III colon cancer, 5-FU vs. 5-FU plus oxaliplatin, n ¼ 892) [18]. QUASAR (QUick and simple and reliable) QUASAR was a large (N ¼ 3239) randomized trial conducted between 1994 and 2003 to determine the survival benefit from adjuvant chemotherapy with 5-FU and LV vs. surgery alone in colon and rectal cancer [19]. From this trial, archival fixed, paraffinembedded, primary colon tumor blocks from 1436 patients with stage II colon cancer (surgery alone, n ¼ 711; surgery plus 5-FU/LV, n ¼ 725) were used in the prospectively designed validation study. The mean age of the population was 62 years (range, 30e83 years); 42% were women. Baseline patient and tumor characteristics were well balanced between the two treatment arms. For each tissue sample, the following measures were assessed centrally through review of H&E slides and pathology reports by pathologists blinded to treatment allocation and patient outcome: tumor site, histology subtype, grade, and T stage; number and status of lymph node samples; and LVI. MMR status was assessed by immunohistochemistry for MLH1 and MSH2 proteins at a single CLIA-certified laboratory [16]. The 12-gene colon cancer assay was then performed, with Recurrence Score results calculated from the prespecified algorithm (Fig. 1). The study demonstrated a significant association between risk of recurrence and continuous Recurrence Score values (P ¼ 0.004) in patients with stage II colon cancer randomized to surgery alone, with higher scores associated with higher recurrence risk (Fig. 2). Recurrence Score results were also significantly associated with disease-free survival (DFS) and OS. It is important to note that the clinical utility of the assay resides in the continuous nature of the Recurrence Score value, which provides 3-year recurrence risk
64
Y.N. You et al. / Surgical Oncology 24 (2015) 61e66
Figure 2. Colon cancer Recurrence Score result predicts recurrence following surgery. The graph shows Cox model estimates of recurrence risk at 3 years (with 95% confidence intervals [CIs]) by continuous Recurrence Score. There was a significant association between risk of recurrence and continuous Recurrence Score values (mean, 33.1; standard deviation [SD], 12.7; interquartile range [IQR], 18.2) in stage II colon cancer patients randomized to surgery alone, with a higher score associated with higher recurrence risk (hazard ratio [HR] per IQR, 1.38; 95% CI, 1.11e1.74; P ¼ 0.004). A rug plot depicting the distribution of recurrence score is included at the bottom of the graph.
estimates over a range from a low of approximately 9% to a high exceeding 25% [16]. In a multivariable analysis controlling for the effects of tumor grade, number of nodes examined, and LVI, the Recurrence Score, MMR status, and T stage were found to be the most significant, independent predictors of recurrence risk following surgery. Similar to previous reports, T4 tumors were associated with higher recurrence risk (HR 1.83, P ¼ 0.005) and MMR-D tumors were associated with a lower risk of recurrence (HR 0.32, P < 0.001) (Table 2) [16]. The Recurrence Score result predicted risk of recurrence beyond conventional markers with the greatest clinical utility in standard-risk patients with T3, MMR-P, stage II colon cancer (~70% of patients). These results support the value of a new, individualized approach to the assessment of risk of recurrence in stage II colon cancer on the basis of the Recurrence Score, MMR status, and T stage. CALGB 9581 (cancer and leukemia group B) study The CALGB 9581 parent study was a randomized trial comparing adjuvant treatment with edrecolomab, an experimental immunotherapeutic agent, against observation only in patients (n ¼ 1713) with resected, stage II colon cancer [20]. Edrecolomab showed no evidence of clinical efficacy. Compared with QUASAR, this trial had more patients who were 70 years of age and older (36%), whose Table 2 Analysis of association of multiple variables with risk of recurrence in patients with stage II colon cancer in QUASAR study (N ¼ 605). Variable
Categories
HR
95% CI
P-value
Recurrence score Mismatch repair T stage Tumor grade Nodes examined Lymphovascular Invasion
Continuous per 25 units Deficient vs proficient T4 vs T3 High vs low <12 vs 12 Present vs absent
1.43 0.36 1.87 0.65 1.43 1.43
(1.11, (0.17, (1.25, (0.42, (0.98, (0.90,
0.006 0.004 0.004 0.050 0.058 0.150
1.83) 0.80) 2.81) 1.02) 2.08) 2.28)
Abbreviations: CI, confidence interval; LVI, lymphovascular invasion; MMR, mismatch repair. Adapted from Gray RG, Quirke P, Handley K et al. J Clin Oncol. 2011; 29(35):4611e9.
tumors were MMR-D (21%), and who had, on average, lower risk of recurrence (approximately 15% at 5 years). For these reasons, archived tumor specimens from this trial provided a unique opportunity to validate the 12-gene colon cancer assay in a relatively low-risk population. Tissue was available for 1137 (68%) of 1672 stage II colon cancer patients from the CALGB 9581 study. The final study population included 690 patients from both study arms, including all available patients with recurrences (162 events) and a random sample of patients without recurrence (1:3 ratio). The Recurrence Score was calculated using the same prespecified gene list and algorithm previously tested in the QUASAR validation study (Fig. 1) [17]. Consistent with QUASAR, in the primary analysis there was a significant association between the Recurrence Score result and recurrence risk (P ¼ 0.013) in patients with stage II colon cancer. A pre-specified multivariate analysis showed that the Recurrence Score result independently predicted recurrence risk (P ¼ 0.004) after controlling for MMR status, T stage, number of nodes examined, histologic grade, and LVI. In a univariate analysis MMR status was the only covariate significantly associated with recurrence free interval (P ¼ 0.04). Even in this relatively low-risk population, the 12-gene assay was able to identify 22% of patients with T3 MMR-P tumors with an average 5-year risk of recurrence of 21% [17]. These results provide further evidence of the ability of the Recurrence Score result to quantify the risk of recurrence in stage II colon cancer and to further discriminate risk level in patients with T3, MMR-P tumors, which may help to identify patients for whom adjuvant chemotherapy may be considered. NSABP C-07 More recently, the performance of the 12-gene colon cancer assay was validated in a prospectively designed study using archived tissue (n ¼ 892) and prespecified endpoints, methods, and analysis from patients with stages II and III colon cancer randomized to 5-FU or 5-FU plus oxaliplatin in the NSABP C-07 study. NSABP C-07 was a randomized phase III clinical trial that evaluated the efficacy and safety of adding oxaliplatin to bolus 5-FU/LV chemotherapy following surgical resection in patients with stage II and III colon cancer [21]. Among the 892 patients evaluated, a total of 264 (29.6%) patients had stage II disease, 409 (45.9%) patients had stage IIIA/B disease, and 219 (24.6%) had stage IIIC disease. In this study, the continuous Recurrence Score result was significantly associated with a risk of recurrence after controlling for the effects of treatment and stage (P < 0.001). The Recurrence Score assay performed similarly in stages II and III colon cancer and was significantly associated with risk of recurrence after adjusting for T and N stage, number of nodes examined, MMR status, histologic grade, and treatment (P < 0.001) [18]. With the relative benefit of oxaliplatin being similar across the range of Recurrence Score values, the absolute benefit of oxaliplatin therefore increased with higher Recurrence Score results. The increase in absolute benefit between patients with high and low Recurrence Score results was most pronounced in stage II and IIIA/B disease. These results extend the findings of the QUASAR and CALGB 9581 studies and further validate the performance of the 12-gene colon cancer assay in stage II and stage III colon cancer patients treated with 5-FU and 5FU þ oxalipaltin. Impact of the 12-gene colon cancer assay on clinical decision making While oxaliplatin/5-FU-based regimens may improve DFS and OS, the toxic effects of such regimens, such as peripheral neuropathy, have been shown to persist as long as 4 years after
Y.N. You et al. / Surgical Oncology 24 (2015) 61e66
treatment [5]. The utility of the Recurrence Score result in clinical practice allows high-risk and low-risk patients to be distinguished so that high-risk patients can receive therapies that offer a larger potential absolute benefit, whereas low-risk patients can be spared unnecessary treatment that may offer only minimal benefit. The Recurrence Score result enables discrimination of absolute benefit as a function of risk. Accordingly, surgeons and medical oncologists can inform their patients with lower Recurrence Score results about the trade-offs between smaller absolute benefits and the risks of chemotherapy-related toxicities, while patients with higher scores may be candidates for chemotherapy treatment as they would be expected to derive greater absolute benefit. The impact of the 12-gene colon cancer assay results on stage II colon cancer in clinical practice was evaluated after the first two years of use in a survey of medical oncologists [22]. Subsequently, two prospective studies examined the impact of the assay on clinical decision making at multiple centers within the Mayo Clinic Cancer Research Consortium network [23], and in patients from the Clalit Health Services in Israel [24]. Three hundred forty-six U.S. medical oncologists who had ordered the 12-gene assay completed a 34-item survey from December 2010 to December 2011. Among 92 physicians who reported pre- and post-assay recommendations based upon chart reviews, 27 (29%) changed their recommendations after receiving the 12-gene colon cancer assay results: 2/3 of the treatment recommendations decreased in intensity, whereas 1/3 increased in intensity. These changes were in agreement with Recurrence Score results, in that treatment intensity decreased more often for lower Recurrence Score values and increased for higher scores (P ¼ 0.0035 for the trend). These results suggest that use of the assay in clinical practice may be associated with meaningful changes in treatment recommendations for patients with stage II colon cancerdspecifically, reductions in treatment intensity that could spare patients unnecessary toxic effects of chemotherapy [22]. These results were later confirmed in the first prospective study to assess the role of the quantitative Recurrence Score result in clinical decision-making for patients with stage IIA, MMR-P colon cancer. This multicenter prospective study was conducted by more than 100 oncologists within the Mayo Clinic Cancer Research Consortium network who enrolled patients with stage IIA (T3N0) colon cancer across 17 academic and community sites in the study to characterize the impact of Recurrence Score results on medical oncologists' recommendations regarding adjuvant chemotherapy in T3, MMR-P, stage II colon cancer patients. Tumor specimens from patients with resected stage IIA colon cancer who were candidates for adjuvant chemotherapy were assessed using the Recurrence Score value and MMR status. Physicians completed surveys prior to and after receiving the Recurrence Score report along with MMR results and recorded their planned treatments. In the primary analysis treatment recommendations changed for 63 (45%) of 141 MMR-P patients with most (3/4) changes in the direction of decreased intensity. Recommendations for chemotherapy decreased from 52% of patients pre-assay to 30% postassay. The direction of change was concordant with the Recurrence Score result (p ¼ 0.011). These findings indicate that use of the 12-gene assay may lead to overall reductions in the use of adjuvant chemotherapy in T3, MMR-P, stage II colon cancer patients [23]. Secondary analyses were carried out to evaluate patient treatment decisions, physician confidence, concordance in treatment choice between physicians and patients, and patient decisional conflict after receiving the 12-gene assay result. One hundred ninety of 221 patients enrolled were evaluable, including 139 with MMR-P tumors. Post-assay treatment decisions in 129 (96%) of 135
65
patients were concordant between patients and physicians compared to 49 (66%) of 74 treatment decisions prior to receiving the 12-gene assay result. In the majority of cases, patients (85%) and physicians (69%) reported that the assay influenced their treatment decisions. Further, the 12-gene assay results increased physician confidence in their treatment recommendations in 126 (84%) and provided additional clinically relevant information to physicians in 129 (86%) of cases. Patient decisional conflict was significantly lower after assay results (p < 0.001) [25]. In the Clalit Health Services study, investigators prospectively collected treatment recommendations before testing with the 12gene assay and actual treatment received following receipt of test results in T3 MMR-P stage II colon cancer patients between January 2011 and May 2012. Pre-assay treatment recommendations were compared with assay results and actual treatment delivered. Preassay treatment recommendations differed from post-assay treatment delivered in 102 (38%) of 269 evaluable patients, with less treatment delivered for 76 (28%) and more for 26 (10%). Pre-assay, 121 (45%) patients received recommendation for chemotherapy; post-assay, 75 (28%) actually received chemotherapy. Changes in treatment intensity were generally consistent with the Recurrence Score results with increased treatment intensity observed more often at higher scores and decreased intensity at lower scores [24]. Consistent with prior results, in this first international study of the role of the 12-gene assay in clinical decision making, the use of the 12-gene assay had an impact on treatment decisions in Israel, with an overall net reduction in chemotherapy use. Economic impact of the 12-gene colon cancer assay in clinical care In 2012 an economic analyses comparing the 12-gene assay to published patterns of care from the NCCN® Colon/Rectum Cancer Outcomes Database predicted that the assay would improve outcomes while lowering medical costs for stage II, T3, MMR proficient patients [26]. Health economic analysis from this study projected that use of the Recurrence Score result to guide treatment decisions could reduce the use of adjuvant chemotherapy by an average of 17%, result in an average decrease in direct medical costs of nearly $3000 per patient, and increase quality-adjusted life expectancy by an average of 0.035 years [26]. A more recent study [26] assessed the validity of these findings using data from the prospective Mayo Clinic Cancer Research Consortium clinical utility study discussed above. Results of this study revealed a decrease in recommendations towards any adjuvant chemotherapy use by 22%, with a decrease in recommendations towards 5-FU þ oxaliplatin adjuvant chemotherapy use of 11%. Quality-adjusted survival increased on average by 0.114 years, and average lifetime medical savings was estimated at $991 per patient [27]. The cost savings to patients was attributed to an anticipated lower total cost of adjuvant chemotherapy resulting from decreases in cost associated with drugs, administration and adverse events management. These savings were expected to persist even if the cost of oxaliplatin dropped by 75% as a result of generic substitution. These analyses confirm the results of the earlier economic analysis and show that use of the 12-gene assay increases quality-adjusted survival and may be economically favorable for patients with stage II, T3, MMR-P colon cancer. Discussion/conclusion Because the role of adjuvant therapy in patients with stage II colon cancer remains controversial, yet the toxicity risk of chemotherapy is well-known and real, the optimal treatment plan should be tailored to each patient's level of individual risk. Surgeons, who provide the mainstay of treatment for patients with stage II and III colon cancer, namely, oncologic resection of tumor, may be involved
66
Y.N. You et al. / Surgical Oncology 24 (2015) 61e66
in multidisciplinary tumor boards with medical oncologists to make recommendations for adjuvant therapy and/or post-resection surveillance. Therefore, there is a critical need for tools that allow surgeons to accurately assess each patient's risk of recurrence and expected absolute benefit from adjuvant chemotherapy. Pathologic tumor stage and MMR status, along with other pathologic features, are conventionally used as the first steps in the risk assessment; but molecular-based diagnostic tests could be used to reflect the underlying tumor biology and provide more precise estimates of recurrence risk. An area of great practical benefit from such as individualized risk assessment assay includes standard-risk patients with T3, MMR-P, stage II colon cancer, for whom conventional prognostic factors are not informative. The 12-gene colon cancer assay in stage II colon cancer exemplifies a tool for personalized risk assessment that can aid the surgeon and medical oncologist in assessing recurrence risk, and deciding about adjuvant chemotherapy, while optimizing the quality-adjusted survival. Another area of clinical benefit is the discrimination of absolute oxaliplatin benefit as a function of risk in patients with stage III colon cancer [18]. Patients with lower Recurrence Score results, particularly those of older age, competing comorbidities, and/or concerns about quality of life, may be counseled about the trade-offs between smaller absolute benefits and the risks of oxaliplatin-related toxicities. In summary, the 12-gene colon cancer assay provides clinical utility to surgeons and medical oncologists making decisions about oxaliplatin and/or 5-FU based adjuvant chemotherapy in patients with stage II and also stage III colon cancer. Conflict of interest statement Y. You, R. Rustin, and J. Sullivan received an honorarium for their time and efforts. Authorship statement Guarantor of the integrity of the study: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Study concepts: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Study design: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Definition of intellectual content: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Literature research: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Clinical studies: Y. Nancy You Data acquisition: Y. Nancy You Data analysis: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Manuscript preparation: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Manuscript editing: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin Manuscript review: Y. Nancy You, James D. Sullivan, Rudolph B. Rustin References [1] Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics. CA Cancer J Clin 2014;2014(64):9e29. [2] NCCN. NCCN clinical practice guidelines in oncology: colon cancer version 2. 2015. ed2015. [3] Sargent DJ, Marsoni S, Monges G, Thibodeau SN, Labianca R, Hamilton SR, et al. Defective mismatch repair as a predictive marker for lack of efficacy of fluorouracil-based adjuvant therapy in colon cancer. J Clin Oncol 2010;28: 3219e26. [4] Midgley R, Rasul K, Al Salama H, Kerr DJ. Gene profiling in early stage disease. Cancer J 2010;16:210e3.
[5] Andre T, Boni C, Navarro M, Tabernero J, Hickish T, Topham C, et al. Improved overall survival with oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment in stage II or III colon cancer in the MOSAIC trial. J Clin Oncol 2009;27:3109e16. [6] Kuebler JP, Colangelo L, O'Connell MJ, Smith RE, Yothers G, Begovic M, et al. Severe enteropathy among patients with stage II/III colon cancer treated on a randomized trial of bolus 5-fluorouracil/leucovorin plus or minus oxaliplatin: a prospective analysis. Cancer 2007;110:1945e50. [7] Land SR, Kopec JA, Cecchini RS, Ganz PA, Wieand HS, Colangelo LH, et al. Neurotoxicity from oxaliplatin combined with weekly bolus fluorouracil and leucovorin as surgical adjuvant chemotherapy for stage II and III colon cancer: NSABP C-07. J Clin Oncol 2007;25:2205e11. [8] NCCN. NCCN clinical practice guidelines in oncology: breast cancer. v3. 2014. ed2014. [9] Harris L, Fritsche H, Mennel R, Norton L, Ravdin P, Taube S, et al. American society of clinical oncology 2007 update of recommendations for the use of tumor markers in breast cancer. J Clin Oncol 2007;25:5287e312. [10] Goldhirsch A, Winer EP, Coates AS, Gelber RD, Piccart-Gebhart M, Thurlimann B, et al. Personalizing the treatment of women with early breast cancer: highlights of the St Gallen International Expert Consensus on the primary therapy of early breast cancer 2013. Ann Oncol 2013;24:2206e23. [11] Gene expression profiling and expanded immunohistochemistry tests for guiding adjuvant chemotherapy decisions in early breast cancer management: MammaPrint, Oncotype DX, IHC4 and Mammostrat. National Institute for Health and Care Excellence; 2013. [12] Clark-Langone KM, Sangli C, Krishnakumar J, Watson D. Translating tumor biology into personalized treatment planning: analytical performance characteristics of the oncotype DX colon cancer assay. BMC Cancer 2010;10:691. [13] Webber EM, Lin JS, Evelyn PW. Oncotype DX tumor gene expression profiling in stage II colon cancer. Application: prognostic, risk prediction. PLoS Curr 2010:2. [14] Clark-Langone KM, Wu JY, Sangli C, Chen A, Snable JL, Nguyen A, et al. Biomarker discovery for colon cancer using a 761 gene RT-PCR assay. BMC Genomics 2007;8:279. [15] O'Connell MJ, Lavery I, Yothers G, Paik S, Clark-Langone KM, Lopatin M, et al. Relationship between tumor gene expression and recurrence in four independent studies of patients with stage II/III colon cancer treated with surgery alone or surgery plus adjuvant fluorouracil plus leucovorin. J Clin Oncol 2010;28:3937e44. [16] Gray RG, Quirke P, Handley K, Lopatin M, Magill L, Baehner FL, et al. Validation study of a quantitative multigene reverse transcriptase-polymerase chain reaction assay for assessment of recurrence risk in patients with stage II colon cancer. J Clin Oncol 2011;29:4611e9. [17] Venook AP, Niedzwiecki D, Lopatin M, Ye X, Lee M, Friedman PN, et al. Biologic determinants of tumor recurrence in stage II colon cancer: validation study of the 12-gene recurrence score in cancer and leukemia group B (CALGB) 9581. J Clin Oncol 2013;31:1775e81. [18] Yothers G, O'Connell MJ, Lee M, Lopatin M, Clark-Langone KM, Millward C, et al. Validation of the 12-gene colon cancer recurrence score in NSABP C-07 as a predictor of recurrence in patients with stage II and III colon cancer treated with fluorouracil and leucovorin (FU/LV) and FU/LV plus oxaliplatin. J Clin Oncol 2013;31:4512e9. [19] Quasar Collaborative G, Gray R, Barnwell J, McConkey C, Hills RK, Williams NS, et al. Adjuvant chemotherapy versus observation in patients with colorectal cancer: a randomised study. Lancet 2007;370:2020e9. [20] Niedzwiecki D, Bertagnolli MM, Warren RS, Compton CC, Kemeny NE, Benson 3rd AB, et al. Documenting the natural history of patients with resected stage II adenocarcinoma of the colon after random assignment to adjuvant treatment with edrecolomab or observation: results from CALGB 9581. J Clin Oncol 2011;29:3146e52. [21] Kuebler JP, Wieand HS, O'Connell MJ, Smith RE, Colangelo LH, Yothers G, et al. Oxaliplatin combined with weekly bolus fluorouracil and leucovorin as surgical adjuvant chemotherapy for stage II and III colon cancer: results from NSABP C-07. J Clin Oncol 2007;25:2198e204. [22] Cartwright T, Chao C, Lee M, Lopatin M, Bentley T, Broder M, et al. Effect of the 12-gene colon cancer assay results on adjuvant treatment recommendations in patients with stage II colon cancer. Curr Med Res Opin 2014;30:321e8. [23] Srivastava G, Renfro LA, Behrens RJ, Lopatin M, Chao C, Soori GS, et al. Prospective multicenter study of the impact of oncotype DX colon cancer assay results on treatment recommendations in stage II colon cancer patients. Oncologist 2014;19:492e7. [24] Brenner B, Lopatin M, Lee M, Geva R, Beny A, Dror Y, et al. Impact of the 12gene colon cancer recurrence score assay on clinical decision-making for adjuvant therapy in stage II colon cancer patients in Israel. In: European Cancer conference. Amsterdam, Netherlands; 2012. [25] Alberts SR, Renfro LA, Lopatin M, Tezcan H, Sloan J, Chao C, et al. Prospective evaluation of a 12-gene assay on patient treatment decisions and physician confidence in mismatch repair-proficient (MMR-P) stage IIa colon cancer patients. In: ASCO gastrointestinal cancers symposium. San Francisco, CA; 2014. [26] Hornberger J, Lyman GH, Chien R, Meropol NJ. A multigene prognostic assay for selection of adjuvant chemotherapy in patients with T3, stage II colon cancer: impact on quality-adjusted life expectancy and costs. Value Health 2012;15:1014e21. [27] Alberts SR, Yu TM, Behrens RJ, Renfro LA, Srivastava G, Soori GS, et al. Comparative economics of a 12-gene assay for predicting risk of recurrence in stage II colon cancer. Pharmacoeconomics 2014;32:1231e43.