Survival of patients with metastatic colorectal cancer has dramatically improved over the past 20 years, primarily because physicians have become adept at using the many regimens approved for this patient population. Future advances may come from understanding molecular subtypes, finding and treating new actionable mutations, and harnessing the immune system.
Axel Grothey and Alan P. Venook
Results from the pivotal IDEA trial, which evaluated 3 versus 6 months of adjuvant oxaliplatin-based chemotherapy, are incorporated into the NCCN Guidelines for Colon Cancer. The guidelines recommend that for patients with low-risk stage III disease, the preferred regimen is CAPEOX for 3 months or FOLFOX for 3 to 6 months. For patients with high-risk stage III disease, the preferred regimen is CAPEOX for 3 to 6 months or FOLFOX for 6 months. In metastatic disease, tumor sidedness should be a consideration when choosing a biologic. For BRAF-mutated disease, several triplets are now recommended options. Importantly, for a subset of patients with metastatic disease, new to the NCCN Guidelines is the incorporation of nivolumab and pembrolizumab as subsequent therapy for those with microsatellite instability–high or mismatch repair–deficient tumors.
Katherine Van Loon and Alan P. Venook
No definitive evidence shows benefit from adjuvant therapy for stage II colon cancer, and its role remains controversial. Although a trend toward improved disease-free survival (DFS) has been reported in subgroup analyses from clinical trials that included patients with stage II disease, time trends for recurrences of stage II disease indicate that DFS is not a reliable surrogate for overall survival (OS). Several clinical trials have been conducted to answer the question of whether adjuvant therapy benefits patients with stage II disease, but none have been adequately powered to detect what would be a small OS benefit. Features that are currently used to assign high risk for recurrence (tumor perforation, lymphovascular invasion, <12 lymph nodes analyzed, and poorly differentiated histology) may or may not be associated with clinical outcome, and they are not predictive of treatment benefit. Risks of adjuvant therapy are nonnegligible and must be weighed against a large number of patients needed to be treated to realize benefit. Future research should aim to answer the questions of whether microsatellite instability, nodal sampling, molecular markers, and genetic signatures are useful tools to guide decision-making. Given what is now known, the viewpoint is that the aggregate data do not support adjuvant therapy for patients with normal-risk stage II colon cancer.
David Y. Oh, Alan P. Venook and Lawrence Fong
Although overall survival from colorectal cancer (CRC) has steadily improved over the past decade, there is still work to be done. The gains associated with improved detection and treatment paradigms with chemotherapy and biologics appear to have reached their ceiling. Immune-based therapies have recently demonstrated clinical benefit in other cancers, including CRC with microsatellite instability (MSI), but patients with CRC without MSI have not yet derived benefit. This article reviews the history of CRC immunotherapy trials, the conceptual basis for why the activity of the immune system may be relevant to survival in CRC, and current efforts in CRC immunotherapy, and speculates about future efforts in this area based on experience with immunotherapy efforts in other classes of solid tumors.
Aparna Parikh, Chloe Atreya, W. Michael Korn and Alan P. Venook
HER2 gene amplifications and activating mutations in the HER2 receptor tyrosine kinase are present in 4% of metastatic colorectal cancers (mCRCs). HER2-targeted therapy is not standard of care, although preclinical and clinical data suggest that patients with HER2 amplifications and/or HER2-activating mutations may benefit from HER2-directed therapy. HER2 amplifications and activating mutations have also been implicated in resistance to anti–epidermal growth factor receptor–based therapy. This report describes a patient with KRAS, NRAS, and BRAF wild-type mCRC who experienced disease progression on first-line treatment with FOLFIRI and cetuximab after only 5 months, and subsequently experienced progression on second-line treatment with capecitabine and oxaliplatin plus bevacizumab after 2 months with significant functional decline. Next-generation sequencing of the primary tumor identified HER2 amplification, and we were able to obtain trastuzumab-DM1 for off-label use. The patient had symptomatic clinical benefit from trastuzumab-DM1 and had radiographic disease control for 7 months. On progression, therapy was changed to trastuzumab and pertuzumab, but the patient's disease progressed 3 months later. Treatment with the trastuzumab-DM1 resulted in a sustained response that was longer than his prior responses in the first and second lines of treatment, with a dramatic improvement in the patient's functional status. This case represents the first report, to our knowledge, of successful single-agent treatment of HER2-amplifed CRC with trastuzumab-DM1. Clinical trials targeting patients with HER2-mutated and -amplified metastatic colon cancer are currently underway. Molecular insights from investigating HER2 activation and the impact of HER2-directed therapies in a wide variety of solid tumors will create the needed evidence base to more broadly inform patient care.
Robin K. Kelley, Grace Wang and Alan P. Venook
Biomarkers reflective of the molecular and genetic heterogeneity in colorectal cancers now guide certain aspects of clinical management and offer great potential for enrichment, stratification, and identification of novel therapeutic targets in drug development. Using case-based examples, this article reviews biomarkers that have an established role in the clinical management of colorectal cancer: mismatch repair protein testing and KRAS and BRAF mutational analysis. A selection of biomarkers undergoing validation for future clinical application is presented, and the dynamic and challenging interface between biomarkers in research and clinical practice is discussed.
Robin K. Kelley, Chloe Atreya, Alan P. Venook and Phillip G. Febbo
Because of a surge in molecular testing capabilities concurrent with the rising numbers of targeted therapies in clinical development, the commercial use of predictive biomarkers before clinical validation is available is a topic of growing relevance to medical oncologists. Increasingly, patients will present questions about, requests for, and results from commercial biomarker tests for their oncologists to address. The sheer numbers of tests reaching the market, along with forecasted American Medical Association reforms in current procedural terminology coding and increasing FDA oversight of in vitro companion diagnostic device development, are likely to draw intense scrutiny to the regulation of commercial molecular testing in the near future, which will also require clinicians to remain abreast of the level of clinical validation of the biomarker tests available in practice. In addition to the direct risks of novel biomarker testing, including financial cost and ethical issues, the indirect risks encompass those associated with any clinical decision based on the biomarker test results. A great need exists for comprehensive and dynamic practice guidelines for all types of biomarker testing according to tumor type.
Robin K. Kelley, Stephanie L. Van Bebber, Kathryn A. Phillips and Alan P. Venook
Predictive and prognostic biomarkers offer a potential means to personalize cancer medicine, although many reach the market-place before they have been validated, and their adoption is often hindered by variable clinical evidence. Because of this variability in supporting evidence, clinical practice guidelines formulated by panels of subspecialty experts may be particularly important in guiding stakeholders' acceptance and use of new personalized medicine biomarker tests and other nascent technologies. This article provides a structured review of the clinical evidence supporting 4 contemporary biomarker tests in colorectal cancer: K-ras and B-raf mutation analyses, mismatch repair protein testing, and the Oncotype DX Colon Cancer Assay. All 4 tests have been evaluated for guideline inclusion by the NCCN Guidelines Panel for Colon Cancer. This case study shows significant variability in the level of clinical evidence associated with these tests. In the cases of B-raf and mismatch repair protein testing, the available evidence is also inconsistent as it pertains to the specific NCCN Guideline recommendation. Based on this uncertainty in the evidence base, the authors conclude that expert clinical judgment, experience, and consensus may be more heavily weighted than published clinical trial data in the evaluation of new personalized medicine biomarker tests. Potential implications of this conclusion and future directions for research are discussed.
Robert W. Carlson, Jonathan K. Larsen, Joan McClure, C. Lyn Fitzgerald, Alan P. Venook, Al B. Benson III and Benjamin O. Anderson
The NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) are evidence- and consensus-based clinical practice guidelines addressing malignancies that affect more than 97% of all patients with cancer in the United States. The NCCN Guidelines are used extensively in the United States and globally. Use of the guidelines outside the United States has driven the need to adapt the guidelines based on local, regional, or national resources. The NCCN Guidelines Panels created, vetted, and continually update the NCCN Guidelines based on published scientific data on cancer detection, diagnosis, and treatment efficacy. The guidelines are developed within the context of commonly available resources, methods of payment, societal and cultural expectations, and governmental regulations as they exist in the United States. Although many of the cancer management recommendations contained in the NCCN Guidelines apply broadly from a global perspective, not all do. Disparities in availability and access to health care exist among countries, within countries, and among different social groups in the same country, especially regarding resources for cancer prevention, early detection, and treatment. In addition, different drug approval and payment processes result in regional variation in availability of and access to cancer treatment, especially highly expensive agents and radiation therapy. Differences in cancer risk, predictive biomarker expression, and pharmacogenetics exist across ethnic and racial groups, and therefore across geographic locations. Cultural and societal expectations and requirements may also require modification of NCCN Guidelines for use outside the United States. This article describes the adaptation process, using the recent Latin American adaptation of the 2013 NCCN Guidelines for Colorectal Cancer as an example.
Chloe E. Atreya, Claire Greene, Ryan M. McWhirter, Nabia S. Ikram, I. Elaine Allen, Katherine Van Loon, Alan P. Venook, Benjamin M. Yeh and Spencer C. Behr
Background: BRAF-mutant metastatic colorectal cancers (mCRCs) share many clinicopathologic features with right-sided colon tumors, including frequent peritoneal involvement. Because of the poorer outcomes associated with BRAF mutations, early enrollment in clinical trials has been encouraged. However, the use of standard eligibility and assessment criteria, such as measurable disease, has anecdotally impeded patient accrual and restricted appraisal of treatment response. We investigated whether the presence of a BRAF V600E mutation is differentially associated with sites and appearance of metastatic disease in patients matched by primary tumor location. Methods: A total of 40 patients with BRAF-mutant mCRC were matched to 80 patients with BRAF wild-type mCRC by location of primary tumor (right or left colon; rectum), sex, and age. Associations between BRAF mutation status and clinicopathologic characteristics and metastatic sites were analyzed using proportion tests. Survival was summarized with Kaplan-Meier and Cox regression methods. Results: The distribution of primary tumor locations was: 60% right colon, 30% left colon, and 10% rectum. Compared with BRAF wild-type tumors, BRAF-mutant tumors more commonly associated with peritoneal metastases (50% vs 31%; P=.045) and ascites (50% vs 24%; P=.0038). In patients with left colon primaries, BRAF mutations were associated with more frequent ascites (58% vs 12%; P=.0038) and less frequent liver metastases (42% vs 79%; P=.024). Among patients with right colon primaries, no significant difference in sites of disease by BRAF mutation status was observed. Disease was not measurable by RECIST 1.1 in 24% of patients with right-sided primary tumors, irrespective of BRAF mutation status. In the BRAF-mutated cohort, ascites correlated unfavorably with survival (hazard ratio, 2.35; 95% CI, 1.14, 4.83; P=.02). Conclusions: Greater frequency of ascites and peritoneal metastases, which pose challenges for RECIST 1.1 interpretation of therapeutic outcomes, are seen with BRAF-mutant mCRC, even when patients are matched for primary tumor location.