Treatment Trends and Clinical Outcomes of Left-Sided RAS/RAF Wild-Type Metastatic Colorectal Cancer in the United States

Authors:
Christopher Nevala-PlagemannDivision of Medical Oncology, Huntsman Cancer Institute at the University of Utah; and

Search for other papers by Christopher Nevala-Plagemann in
Current site
Google Scholar
PubMedClose
 MD
,
Siddharth IyengarDepartment of Internal Medicine, and

Search for other papers by Siddharth Iyengar in
Current site
Google Scholar
PubMedClose
 MD
,
Andrew D. TrunkDepartment of Internal Medicine, and

Search for other papers by Andrew D. Trunk in
Current site
Google Scholar
PubMedClose
 MD
,
Lisa PappasDepartment of Population Health Sciences, University of Utah, Salt Lake City, Utah.

Search for other papers by Lisa Pappas in
Current site
Google Scholar
PubMedClose
 MS
,
Benjamin HaalandDepartment of Population Health Sciences, University of Utah, Salt Lake City, Utah.

Search for other papers by Benjamin Haaland in
Current site
Google Scholar
PubMedClose
 PhD
, and
Ignacio Garrido-LagunaDivision of Medical Oncology, Huntsman Cancer Institute at the University of Utah; and

Search for other papers by Ignacio Garrido-Laguna in
Current site
Google Scholar
PubMedClose
 MD, PhD
View More View Less
Volume/Issue:
Volume 20: Issue 3
Online Publication Date:
04 Feb 2022
DOI:
https://doi.org/10.6004/jnccn.2021.7079
Full access

Background: Post hoc analysis of the CALGB/SWOG 80405 trial suggests that anti-EGFR therapy may be superior to bevacizumab when added to first-line chemotherapy in patients with metastatic colorectal cancer (mCRC) who have left-sided primary tumors. We evaluated trends in use of anti-EGFR agents in patients with left-sided RAS/RAF wild-type (WT) mCRC and compared clinical outcomes among the most commonly used treatment strategies. Methods: A nationwide electronic health record (EHR)–derived deidentified database was reviewed for patients with left-sided RAS/RAF WT mCRC. Treatment trends over time were assessed by fitting a linear model to the percentage of patients receiving anti-EGFR therapy. A propensity score weighted Cox model was used to compare overall survival (OS) stratified by first-line targeted therapy received. Results: A total of 1,607 patients with left-sided RAS/RAF WT mCRC received standard first-line chemotherapy. Of these, 965 (60%) received bevacizumab and 186 (12%) received an anti-EGFR agent. The percentage of patients receiving an anti-EGFR increased from 9% in 2013 to 16% in 2018. Median OS for patients treated with chemotherapy alone was 27.3 months (95% CI, 24.8–32.3), 27.5 months with bevacizumab (95% CI, 25.8–28.9; hazard ratio [HR], 0.88; P=.33), and 42.9 months with an anti-EGFR agent (95% CI, 36.0 to not reached; HR, 0.52; P=.005). Conclusions: This analysis suggests that chemotherapy with bevacizumab remained the most widely used first-line treatment strategy for patients with left-sided RAS/RAF WT mCRC in the United States in 2018. Despite this preference, treatment with an anti-EGFR agent was associated with improved OS.

Background

In 2020, there were an estimated 147,000 new diagnoses of colorectal cancer (CRC) in the United States, with an estimated 53,000 deaths.1 Although these numbers are still unacceptable, mortality related to CRC has been declining over the past several decades. This decrease is likely attributable to improved detection of early-stage disease through increased use of appropriate cancer screening and improvements in cytotoxic and targeted therapy. Unfortunately, outcomes for patients who develop or present with metastatic CRC (mCRC) remain poor.

Several targeted therapies have been approved by the FDA for use in the management of RAS/RAF wild-type (WT) mCRC. These agents include bevacizumab, a monoclonal antibody targeting VEGF, and the anti-EGFR antibodies cetuximab and panitumumab.2–4 All 3 of these agents are now approved for use in the first-line setting in patients with RAS/RAF WT mCRC.5–7

Increased use of these agents has likely led to improvement in the survival of patients with mCRC; however, many questions remain regarding the use of targeted therapy in this disease. The phase III FIRE-3 and CALGB/SWOG 80405 trials were designed to identify an optimal first-line treatment strategy by comparing outcomes of patients treated with first-line bevacizumab versus an anti-EGFR agent. Unfortunately, the initial results of these trials were conflicting.8,9 As a result of this uncertainty, the current NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines; Version 2.2021) do not specify a preferred chemotherapy plus targeted therapy combination for the up-front management of fit patients with left-sided RAS/RAF WT mCRC.10

Clearly there is a need to identify patient or tumor characteristics that may predict response to targeted therapy in this population of patients. A promising candidate for this role is primary tumor sidedness. Retrospective analyses of the FIRE-3 and SWOG 80405 trials have suggested that survival outcomes in patients with left-sided primary tumors (LSPTs) treated with an anti-EGFR agent may be superior compared with patients treated with bevacizumab in the first-line setting.11,12 This finding has been supported by small retrospective studies.13,14

In this study, we assess how US clinicians are using targeted therapy in the management of patients with left-sided RAS/RAF WT mCRC. We hypothesized that the relative use of anti-EGFR agents compared with bevacizumab would have increased over the past decade. We also hypothesized that our analysis of a large real-world population of patients with mCRC would similarly suggest superiority of anti-EGFR agents over bevacizumab in those with left-sided RAS/RAF WT tumors.

Methods

Data Source

Data for this study were obtained via review of the nationwide, deidentified electronic health record (EHR)–derived Flatiron Health database. During the study period, the deidentified data originated from approximately 280 US cancer clinics (∼800 sites of care). Most patients in the database originate from community oncology settings; however, the relative community/academic proportions may vary depending on the study cohort. Patient-level deidentified data in the longitudinal Flatiron Health database includes both structured and unstructured data curated using a technology-enabled abstraction process.15,16 The data are deidentified and subject to obligations to prevent reidentification and protect patient confidentiality. Institutional Review Board approval of the study protocol was obtained prior to study conduct and included a waiver of informed consent.

Study Population

The deidentified EHR-derived database was reviewed for patients diagnosed with mCRC between January 2013 and December 2018. Within this cohort, we selected patients who received a standard first-line chemotherapy doublet defined as 5-FU/oxaliplatin (FOLFOX), capecitabine/oxaliplatin (CAPEOX), or 5-FU/irinotecan (FOLFIRI). Within this group, only patients who were documented as having RAS/RAF WT CRC with an LSPT as determined by ICD coding were included in our final analysis. Patients were excluded from analysis if they did not have a visit or medication order within 90 days of metastatic diagnosis to ensure that all patients included in the analysis were engaged with care at the data-providing institution. Targeted therapy (bevacizumab, cetuximab, or panitumumab) was considered part of first-line therapy if it was added within 2 months of the first administered dose of chemotherapy.

Statistical Analysis

Statistical differences between groups stratified by first-line targeted therapy received (none, anti-EGFR, or bevacizumab) were determined using chi-square tests for categorical variables that included race, gender, ECOG performance status prior to first-line therapy (0, 1, 2, or 3–4), first-line chemotherapy backbone received (FOLFOX, FOLFIRI, or CAPEOX), stage at initial diagnosis (de novo or recurrent metastatic disease), microsatellite instability (MSI) and/or mismatch repair (MMR) status recorded within 6 months of metastatic diagnosis (proficient, deficient, unknown), and primary tumor location (rectum, sigmoid, descending, or splenic flexure), and a Wilcoxon test was used to compare age at the time of metastatic diagnosis. The percentage of patients with missing data for any of these variables were reported and these patients were included in all subsequent analyses using “missing” as the value for the respective variable. Trends in the use of first-line targeted therapy over time were assessed by fitting a linear model to the percentage of patients receiving first-line anti-EGFR therapy plotted by quarter.

Survival outcomes between groups were compared using the Kaplan-Meier method and Cox proportional hazard modeling. Overall survival (OS) was defined as the time in months from the date of mCRC diagnosis to the date of death. Patients were censored based on the date of last confirmed activity. Multivariable models were adjusted for chemotherapy backbone (FOLFOX, CAPEOX, FOLFIRI), LSPT location (splenic flexure, descending, sigmoid, rectum), age, gender, stage at initial diagnosis, MSI/MMR status, and ECOG performance status (0, 1, 2, 3–4).

A propensity score weighted survival analysis was used to reduce confounding further when comparing outcomes between groups. Propensity scores were based on a random forest model that included covariates potentially associated with receiving bevacizumab or anti-EGFR treatment.17 These covariates included age at metastatic diagnosis, gender, race, year of metastatic diagnosis, geographic location, ECOG performance status, body mass index, albumin level, CEA level, chemotherapy backbone, presence of hypertension, and prior receipt of anticoagulation therapy. For variables with multiple values over time, the value recorded closest to and before the initiation of first-line therapy was used. An extension of the inverse probability of treatment weighting was used to reweight treatment groups based on these variables. Statistical analysis was performed using R version 3.5.2 (R Foundation for Statistical Computing).

Results

Patient Characteristics

Of 20,333 patients with mCRC included in the EHR-derived dataset, 9,753 received a standard chemotherapy doublet with or without targeted therapy and had a visit or medication order within 90 days of metastatic diagnosis. After exclusion of those with RAS/RAF-mutant or unknown mutational status, 5,876 patients were available for additional analysis. Within this group, 2,449 (41%) had a tumor sidedness–specific ICD code, with 1,607 (65%) documented as having an LSPT. Within this cohort, 456 patients (28%) received first-line chemotherapy alone, 965 (60%) received chemotherapy plus bevacizumab, and 186 (12%) received chemotherapy plus an anti-EGFR agent (Figure 1). When stratified by first-line therapy received, groups were balanced with regard to race, gender, performance status, disease presentation (recurrent vs de novo metastatic disease), and documented location of primary tumor. A statistically significant difference between treatment groups was identified with regard to age, MMR status, and chemotherapy backbone used (Table 1).

Figure 1.
Figure 1.

Patient cohort selection flowchart.

Abbreviations: 1L, first-line; mCRC, metastatic colorectal cancer.

Citation: Journal of the National Comprehensive Cancer Network 20, 3; 10.6004/jnccn.2021.7079

Table 1.

Patient Characteristics Stratified by First-Line Therapy
Table 1.

Treatment Trends

The relative proportion of patients receiving chemotherapy plus an anti-EGFR agent compared with those who received chemotherapy alone or chemotherapy plus bevacizumab increased from 9% in 2013% to 16% in 2018 (Figure 2). Trends in the use of anti-EGFR therapy over time were assessed by fitting a linear model to quarterly percentages (R2=0.2203). This model estimated an annual increase in the relative use of anti-EGFR therapy of 0.92% (P=.021).

Figure 2.
Figure 2.

Trends in the use of first-line targeted therapies in combination with chemotherapy in patients with metastatic colorectal cancer in the United States from 2013 to 2018.

Abbreviation: chemo, chemotherapy.

Citation: Journal of the National Comprehensive Cancer Network 20, 3; 10.6004/jnccn.2021.7079

Survival Analysis

Median OS for patients treated with an anti-EGFR agent was 42.9 months (95% CI, 36.0 to not reached) compared with 27.5 months (95% CI, 25.8–28.9) for those treated with bevacizumab, and 27.3 months (95% CI, 24.8–32.3) for those treated with chemotherapy alone (Figure 3). Multivariate analysis suggested a significant improvement in OS with the addition of an anti-EGFR agent (hazard ratio [HR], 0.52; 95% CI, 0.33–0.82; P=.005) but not bevacizumab (HR, 0.88; 95% CI, 0.68–1.14; P=.33) compared with chemotherapy alone. A propensity score weighted survival analysis demonstrated similar results, with a significant improvement in OS with the addition of an anti-EGFR agent (HR, 0.56; 95% CI, 0.39–0.81; P=.002) but not bevacizumab (HR, 1.05; 95% CI, 0.88–1.25; P=.605) when compared with chemotherapy alone. In a separate propensity score weighted analysis excluding patients who did not receive targeted therapy, OS was significantly greater among those treated with an anti-EGFR agent compared with those treated with bevacizumab (HR, 0.52; 95% CI, 0.37–0.73; P =.0001). Survival outcomes are summarized in Table 2. Results of an unadjusted exploratory subgroup analysis comparing outcomes of patients treated with an anti-EGFR agent versus bevacizumab stratified by age, gender, chemotherapy backbone, performance status, MMR status, disease presentation, and LSPT location are shown in Figure 4.

Figure 3.
Figure 3.

Kaplan-Meier plot comparing overall survival of patients with left-sided RAS/RAF WT mCRC treated with first-line chemotherapy alone, chemotherapy plus bevacizumab, or chemotherapy plus an anti-EGFR agent.

Abbreviations: chemo, chemotherapy; mCRC, metastatic colorectal cancer; WT, wild-type.

Citation: Journal of the National Comprehensive Cancer Network 20, 3; 10.6004/jnccn.2021.7079

Figure 4.
Figure 4.

Subgroup analysis comparing HRs for the risk of death in patients with left-sided RAS/RAF WT mCRC treated with first-line chemotherapy plus an anti-VEGF agent (bevacizumab) versus chemotherapy plus an anti-EGFR agent (cetuximab or panitumumab).

Abbreviations: CAPEOX, capecitabine/oxaliplatin; EGFR, epidermal growth factor receptor; FOLFIRI, 5-FU/irinotecan; FOLFOX, 5-FU/oxaliplatin; HR, hazard ratio; mCRC, metastatic colorectal cancer; PS, performance status; VEGF, vascular endothelial growth factor; WT, wild type.

Citation: Journal of the National Comprehensive Cancer Network 20, 3; 10.6004/jnccn.2021.7079

Table 2.

Survival Outcomes of Patients With Left-Sided RAS/RAF WT mCRC
Table 2.

Discussion

This analysis of real-world data collected from a large US population adds to a growing list of recent publications that suggest treatment with an anti-EGFR agent in the first-line setting may lead to superior survival outcomes compared with treatment with bevacizumab in patients with left-sided, RAS/RAF WT mCRC. Current clinical guidelines, however, remain ambiguous with regard to optimal treatment of these patients. This is largely because of the conflicting results of the FIRE-3 and CALGB/SWOG 80405 trials, the only large phase III trials to directly compare survival of patients with mCRC treated with a first-line chemotherapy doublet plus either bevacizumab or cetuximab.9,18

Since the publication of these trials, several attempts have been undertaken to explain the discrepancies between their results. Hypotheses proposed include differences in the study population (a mostly US cohort in the SWOG 80405 trial vs a European cohort in FIRE-3), chemotherapy backbone (majority FOLFOX in SWOG 80405 vs only FOLFIRI in FIRE-3), sequence of therapy and subsequent treatment beyond progression, and variation in the proportion of patients with left-sided versus right-sided tumors. Although the debate is ongoing, tumor sidedness appears to be one area in which there is clearly agreement between these trials. Post hoc analyses of both have demonstrated a significantly greater median OS in patients with an LSPT who received cetuximab compared with bevacizumab, with a clinically significant difference of 5.5 months in SWOG 80405 and 12 months in FIRE-3.12,19,20

Given the ease with which tumor sidedness can be assessed and the magnitude of clinical benefit suggested by these post hoc analyses, we hypothesized that their publication would quickly lead to an increase in the use of anti-EGFR agents in patients with left-sided RAS/RAF WT mCRC in the United States. However, our analysis shows that use of first-line anti-EGFR agents in this patient population increased by only 7% from 2013 to 2018, with FOLFOX + bevacizumab remaining the most widely used first-line treatment strategy. Despite this apparent preference for bevacizumab, our survival analysis was consistent with the SWOG 80405 and FIRE-3 studies by demonstrating a >1-year difference in median OS in favor of patients treated with an anti-EGFR agent. This finding was sustained when controlling for relevant patient and tumor characteristics in both our multivariable and propensity score weighted analyses. Furthermore, the results of this study are in agreement with the observations of several other recent retrospective analyses.13,14,21,22

In light of mounting evidence for the inferiority of first-line bevacizumab in left-sided RAS/RAF WT mCRC, it is unclear why uptake of anti-EGFR agents remains low in the United States. Route of administration and dosing schedule are similar, and compared with bevacizumab, which is associated with serious adverse events such hypertension, gastrointestinal perforation, and thromboembolic events, one could argue that anti-EGFR agents are actually less toxic. One exception to this, and a potential explanation for decreased use of anti-EGFR agents, is the acneiform rash associated with this class of medications. It should be noted, however, that prophylaxis with doxycycline has been shown to be effective at preventing these cutaneous adverse effects.23 It is also possible that our analysis, which only included data up until 2018, may not have been able to capture more recent changes in practice patterns influenced by the publication the post hoc analyses of SWOG 80405 and FIRE-3 in 2016 and 2017.

Although the exact mechanisms underlying the impact of tumor sidedness as a prognostic or predictive marker are still being elucidated, it is likely that tumor sidedness is a surrogate for underlying disease biology. This may be explained by differences in the composition of consensus molecular subtypes (CMS), a classification system based on the genetic makeup of the primary tumor and cells in the tumor microenvironment,24 seen in left-sided versus right-sided tumors. CMS2 (canonical/WNT) and CMS4 (mesenchymal) subtypes, found more frequently in LSPTs, have been shown to be more likely to respond to cetuximab than CMS1 (MSI immune) or CMS3 (metabolic) subtypes.25 Indeed, exploratory analyses evaluating the impact of CMS subtypes in SWOG 80405 demonstrated the longest OS in patients with CMS2 treated with cetuximab at 42 months.26

Although tumor sidedness as a surrogate for underlying tumor biology as represented by CMS subtype appears to be a leading hypothesis, other potential explanations for differences in response to targeted therapy based on tumor sidedness include differences in the composition of the gut microbiome, in the typical metastasis pattern of left-sided versus right-sided tumors, and/or in underlying tumor burden.

Another key question in this patient population is whether treatment outcome is influenced not only by the choice of targeted therapy in the first-line setting but also by the selection of targeted therapy/chemotherapy combination. Indeed, this has been proposed as a potential way to explain the contradictory findings of SWOG 80405 and FIRE-3 that differed with regard to chemotherapy backbone used.27 However, our study appears to contradict this hypothesis, as our primary analysis controlled for chemotherapy backbone received and our subgroup analysis showed a trend toward benefit with use of anti-EGFR over bevacizumab regardless of whether FOLFOX or FOLFIRI was used as the chemotherapy backbone.

This study is limited by the retrospective nature of our analysis. Attempts were made to control for confounding using a multivariable and propensity score weighted analysis; however, the possibility of residual confounding cannot be excluded. The use of ICD coding to determine primary tumor sidedness in the Flatiron Health database has been validated; however, this method resulted in exclusion of many patients without a side-specific ICD code.28 Additional limitations include missingness in several covariates and treatment data. Finally, our data only includes patients treated until 2018 and thus may not have captured more recent changes in treatment trends that occurred once anti-EGFR was listed as an option for patients with left-sided tumors by the NCCN Guidelines. Despite these limitations, this study of real-world outcomes in a large, diverse population of patients across the United States demonstrated a statistically and clinically significant difference in outcomes of patients with left-sided RAS/RAF WT mCRC when stratifying by first-line targeted therapy received.

Conclusions

This analysis of real-world data in the United States did not demonstrate a significant increase in the relative percentage of patients with left-sided RAS/RAF WT mCRC being treated with anti-EGFR agents from 2013 to 2018. At the time of data collection, chemotherapy with bevacizumab remained the most widely used first-line treatment strategy for these patients in the United States. Despite this preference, our analysis suggests that survival outcomes may be superior with the addition of a first-line anti-EGFR agent in patients with left-sided RAS/RAF WT mCRC. This study adds to the growing list of recent studies demonstrating superiority of anti-EGFR therapy over bevacizumab in this patient population.

References

  • 1.↑

    Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin 2019;69:7–34.

  • 2.↑

    Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004;350:2335–2342.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3.↑

    Cunningham D, Humblet Y, Siena S, et al. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J Med 2004;351:337–345.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4.↑

    Van Cutsem E, Peeters M, Siena S, et al. Open-label phase III trial of panitumumab plus best supportive care compared with best supportive care alone in patients with chemotherapy-refractory metastatic colorectal cancer. J Clin Oncol 2007;25:1658–1664.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5.↑

    Price TJ, Peeters M, Kim TW, et al. Panitumumab versus cetuximab in patients with chemotherapy-refractory wild-type KRAS exon 2 metastatic colorectal cancer (ASPECCT): a randomised, multicentre, open-label, non-inferiority phase 3 study. Lancet Oncol 2014;15:569–579.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6.↑

    Van Cutsem E, Köhne CH, Hitre E, et al. Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer. N Engl J Med 2009;360:1408–1417.

  • 7.↑

    Douillard JY, Siena S, Cassidy J, et al. Randomized, phase III trial of panitumumab with infusional fluorouracil, leucovorin, and oxaliplatin (FOLFOX4) versus FOLFOX4 alone as first-line treatment in patients with previously untreated metastatic colorectal cancer: the PRIME study. J Clin Oncol 2010;28:4697–4705.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8.↑

    Heinemann V, von Weikersthal LF, Decker T, et al. FOLFIRI plus cetuximab versus FOLFIRI plus bevacizumab as first-line treatment for patients with metastatic colorectal cancer (FIRE-3): a randomised, open-label, phase 3 trial. Lancet Oncol 2014;15:1065–1075.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9.↑

    Venook AP, Niedzwiecki D, Lenz HJ, et al. Effect of first-line chemotherapy combined with cetuximab or bevacizumab on overall survival in patients with KRAS wild-type advanced or metastatic colorectal cancer: a randomized clinical trial. JAMA 2017;317:2392–2401.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 10.↑

    Benson AB, Venook AP, Al-Hawary MM, et al. NCCN Clinical Practice Guidelines in Oncology: Colon Cancer. Version 2.2021. Accessed June 20, 2021. To view the most recent version, visit NCCN.org

    • Search Google Scholar
    • Export Citation
  • 11.↑

    Holch JW, Ricard I, Stintzing S, et al. The relevance of primary tumour location in patients with metastatic colorectal cancer: a meta-analysis of first-line clinical trials. Eur J Cancer 2017;70:87–98.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12.↑

    Tejpar S, Stintzing S, Ciardiello F, et al. Prognostic and predictive relevance of primary tumor location in patients with RAS wild-type metastatic colorectal cancer: retrospective analyses of the CRYSTAL and FIRE-3 trials. JAMA Oncol 2017;3:194–201.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13.↑

    Moretto R, Cremolini C, Rossini D, et al. Location of primary tumor and benefit from anti-epidermal growth factor receptor monoclonal antibodies in patients with RAS and BRAF wild-type metastatic colorectal cancer. Oncologist 2016;21:988–994.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14.↑

    Aljehani MA, Morgan JW, Guthrie LA, et al. Association of primary tumor site with mortality in patients receiving bevacizumab and cetuximab for metastatic colorectal cancer. JAMA Surg 2018;153:60–67.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15.↑

    Ma X, Long L, Moon S, et al. Comparison of population characteristics in real-world clinical oncology databases in the US: Flatiron Health, SEER, and NPCR. medRxiv. Preprint posted online March 18, 2020.

    • Search Google Scholar
    • Export Citation
  • 16.↑

    Birnbaum B, Nussbaum N, Seidl-Rathkopf K, et al. Model-assisted cohort selection with bias analysis for generating large-scale cohorts from the EHR for oncology research. Cornell University Library. January 30, 2020. Accessed December 30, 2020. Available at: https://arxiv.org/abs/2001.09765

    • Search Google Scholar
    • Export Citation
  • 17.↑

    Breiman L. Random forests. Mach Learn 2001;45:5–32.

  • 18.

    Heinemann V, von Weikersthal LF, Decker T, et al. FOLFIRI plus cetuximab or bevacizumab for advanced colorectal cancer: final survival and per-protocol analysis of FIRE-3, a randomized clinical trial. Br J Cancer 2021;124:587–594.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19.↑

    Venook AP, Niedzwiecki D, Innocenti F, et al. Impact of primary (1°) tumor location on overall survival (OS) and progression-free survival (PFS) in patients (pts) with metastatic colorectal cancer (mCRC): analysis of CALGB/SWOG 80405 (Alliance) [abstract]. J Clin Oncol 2016;34(Suppl):Abstract 3504.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20.↑

    Venook AP, Ou FS, Lenz HJ, et al. Primary (1°) tumor location as an independent prognostic marker from molecular features for overall survival (OS) in patients (pts) with metastatic colorectal cancer (mCRC): analysis of CALGB/SWOG 80405 (Alliance) [abstract]. J Clin Oncol 2017;35(Suppl):Abstract 3503.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21.↑

    Grassadonia A, Di Marino P, Ficorella C, et al. Impact of primary tumor location in patients with RAS wild-type metastatic colon cancer treated with first-line chemotherapy plus anti-EGFR or anti-VEGF monoclonal antibodies: a retrospective multicenter study. J Cancer 2019;10: 5926–5934.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22.↑

    Fiala O, Ostasov P, Hosek P, et al. The predictive role of primary tumour sidedness in metastatic colorectal cancer treated with targeted agents. Anticancer Res 2019;39:5645–5652.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23.↑

    Lacouture ME, Mitchell EP, Piperdi B, et al. Skin toxicity evaluation protocol with panitumumab (STEPP), a phase II, open-label, randomized trial evaluating the impact of a pre-emptive skin treatment regimen on skin toxicities and quality of life in patients with metastatic colorectal cancer. J Clin Oncol 2010;28:1351–1357.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24.↑

    Guinney J, Dienstmann R, Wang X, et al. The consensus molecular subtypes of colorectal cancer. Nat Med 2015;21:1350–1356.

  • 25.↑

    Fontana E, Nyamundanda G, Cunningham D, et al. Molecular subtype assay to reveal anti-EGFR response sub-clones in colorectal cancer (CRC) [abstract]. J Clin Oncol 2018;36(Suppl):Abstract 658.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 26.↑

    Lenz HJ, Ou FS, Venook AP, et al. Impact of consensus molecular subtyping (CMS) on overall survival (OS) and progression free survival (PFS) in patients (pts) with metastatic colorectal cancer (mCRC): analysis of CALGB/SWOG 80405 (Alliance) [abstract]. J Clin Oncol 2017;35(Suppl):Abstract 3511.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27.↑

    Aderka D, Stintzing S, Heinemann V. Explaining the unexplainable: discrepancies in results from the CALGB/SWOG 80405 and FIRE-3 studies. Lancet Oncol 2019;20:e274–283.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28.↑

    Luhn P, Kuk D, Carrigan G, et al. Validation of diagnosis codes to identify side of colon in an electronic health record registry. BMC Med Res Methodol 2019;19:177.

    • Crossref
    • Search Google Scholar
    • Export Citation

Submitted March 26, 2021; final revision received June 29, 2021; accepted for publication July 1, 2021. Published online February 4, 2022.

Author contributions: Study concept and design: All authors. Administrative support: Nevala-Plagemann, Garrido-Laguna. Provision of study materials or patients: Nevala-Plagemann. Data acquisition: Nevala-Plagemann, Pappas, Haaland. Data analysis and interpretation: All authors. Manuscript preparation: All authors.

Disclosures: Dr. Haaland has disclosed receiving travel funds from Flatiron Health. The remaining authors have disclosed that they have not received any financial consideration from any person or organization to support the preparation, analysis, results, or discussion of this article.

Funding: Research reported in this publication was supported by the NCI of the NIH under award number P30CA042014-23.

Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Correspondence: Ignacio Garrido-Laguna, MD, PhD, Division of Medical Oncology, Huntsman Cancer Institute, 2000 Circle of Hope, Suite 5507, Salt Lake City, UT 84112. Email: Ignacio.Garrido-Laguna@hci.utah.edu
  • Collapse
  • Expand
  • Figure 1.
    View in gallery
    Figure 1.

    Patient cohort selection flowchart.

    Abbreviations: 1L, first-line; mCRC, metastatic colorectal cancer.

  • Figure 2.
    View in gallery
    Figure 2.

    Trends in the use of first-line targeted therapies in combination with chemotherapy in patients with metastatic colorectal cancer in the United States from 2013 to 2018.

    Abbreviation: chemo, chemotherapy.

  • Figure 3.
    View in gallery
    Figure 3.

    Kaplan-Meier plot comparing overall survival of patients with left-sided RAS/RAF WT mCRC treated with first-line chemotherapy alone, chemotherapy plus bevacizumab, or chemotherapy plus an anti-EGFR agent.

    Abbreviations: chemo, chemotherapy; mCRC, metastatic colorectal cancer; WT, wild-type.

  • Figure 4.
    View in gallery
    Figure 4.

    Subgroup analysis comparing HRs for the risk of death in patients with left-sided RAS/RAF WT mCRC treated with first-line chemotherapy plus an anti-VEGF agent (bevacizumab) versus chemotherapy plus an anti-EGFR agent (cetuximab or panitumumab).

    Abbreviations: CAPEOX, capecitabine/oxaliplatin; EGFR, epidermal growth factor receptor; FOLFIRI, 5-FU/irinotecan; FOLFOX, 5-FU/oxaliplatin; HR, hazard ratio; mCRC, metastatic colorectal cancer; PS, performance status; VEGF, vascular endothelial growth factor; WT, wild type.

Figure 1.

Patient cohort selection flowchart.

Abbreviations: 1L, first-line; mCRC, metastatic colorectal cancer.
Figure 2.

Trends in the use of first-line targeted therapies in combination with chemotherapy in patients with metastatic colorectal cancer in the United States from 2013 to 2018.

Abbreviation: chemo, chemotherapy.
Figure 3.

Kaplan-Meier plot comparing overall survival of patients with left-sided RAS/RAF WT mCRC treated with first-line chemotherapy alone, chemotherapy plus bevacizumab, or chemotherapy plus an anti-EGFR agent.

Abbreviations: chemo, chemotherapy; mCRC, metastatic colorectal cancer; WT, wild-type.
Figure 4.

Subgroup analysis comparing HRs for the risk of death in patients with left-sided RAS/RAF WT mCRC treated with first-line chemotherapy plus an anti-VEGF agent (bevacizumab) versus chemotherapy plus an anti-EGFR agent (cetuximab or panitumumab).

Abbreviations: CAPEOX, capecitabine/oxaliplatin; EGFR, epidermal growth factor receptor; FOLFIRI, 5-FU/irinotecan; FOLFOX, 5-FU/oxaliplatin; HR, hazard ratio; mCRC, metastatic colorectal cancer; PS, performance status; VEGF, vascular endothelial growth factor; WT, wild type.
  • 1.

    Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin 2019;69:7–34.

  • 2.

    Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004;350:2335–2342.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3.

    Cunningham D, Humblet Y, Siena S, et al. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J Med 2004;351:337–345.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4.

    Van Cutsem E, Peeters M, Siena S, et al. Open-label phase III trial of panitumumab plus best supportive care compared with best supportive care alone in patients with chemotherapy-refractory metastatic colorectal cancer. J Clin Oncol 2007;25:1658–1664.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5.

    Price TJ, Peeters M, Kim TW, et al. Panitumumab versus cetuximab in patients with chemotherapy-refractory wild-type KRAS exon 2 metastatic colorectal cancer (ASPECCT): a randomised, multicentre, open-label, non-inferiority phase 3 study. Lancet Oncol 2014;15:569–579.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6.

    Van Cutsem E, Köhne CH, Hitre E, et al. Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer. N Engl J Med 2009;360:1408–1417.

  • 7.

    Douillard JY, Siena S, Cassidy J, et al. Randomized, phase III trial of panitumumab with infusional fluorouracil, leucovorin, and oxaliplatin (FOLFOX4) versus FOLFOX4 alone as first-line treatment in patients with previously untreated metastatic colorectal cancer: the PRIME study. J Clin Oncol 2010;28:4697–4705.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8.

    Heinemann V, von Weikersthal LF, Decker T, et al. FOLFIRI plus cetuximab versus FOLFIRI plus bevacizumab as first-line treatment for patients with metastatic colorectal cancer (FIRE-3): a randomised, open-label, phase 3 trial. Lancet Oncol 2014;15:1065–1075.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9.

    Venook AP, Niedzwiecki D, Lenz HJ, et al. Effect of first-line chemotherapy combined with cetuximab or bevacizumab on overall survival in patients with KRAS wild-type advanced or metastatic colorectal cancer: a randomized clinical trial. JAMA 2017;317:2392–2401.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 10.

    Benson AB, Venook AP, Al-Hawary MM, et al. NCCN Clinical Practice Guidelines in Oncology: Colon Cancer. Version 2.2021. Accessed June 20, 2021. To view the most recent version, visit NCCN.org

    • Search Google Scholar
    • Export Citation
  • 11.

    Holch JW, Ricard I, Stintzing S, et al. The relevance of primary tumour location in patients with metastatic colorectal cancer: a meta-analysis of first-line clinical trials. Eur J Cancer 2017;70:87–98.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12.

    Tejpar S, Stintzing S, Ciardiello F, et al. Prognostic and predictive relevance of primary tumor location in patients with RAS wild-type metastatic colorectal cancer: retrospective analyses of the CRYSTAL and FIRE-3 trials. JAMA Oncol 2017;3:194–201.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13.

    Moretto R, Cremolini C, Rossini D, et al. Location of primary tumor and benefit from anti-epidermal growth factor receptor monoclonal antibodies in patients with RAS and BRAF wild-type metastatic colorectal cancer. Oncologist 2016;21:988–994.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14.

    Aljehani MA, Morgan JW, Guthrie LA, et al. Association of primary tumor site with mortality in patients receiving bevacizumab and cetuximab for metastatic colorectal cancer. JAMA Surg 2018;153:60–67.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15.

    Ma X, Long L, Moon S, et al. Comparison of population characteristics in real-world clinical oncology databases in the US: Flatiron Health, SEER, and NPCR. medRxiv. Preprint posted online March 18, 2020.

    • Search Google Scholar
    • Export Citation
  • 16.

    Birnbaum B, Nussbaum N, Seidl-Rathkopf K, et al. Model-assisted cohort selection with bias analysis for generating large-scale cohorts from the EHR for oncology research. Cornell University Library. January 30, 2020. Accessed December 30, 2020. Available at: https://arxiv.org/abs/2001.09765

    • Search Google Scholar
    • Export Citation
  • 17.

    Breiman L. Random forests. Mach Learn 2001;45:5–32.

  • 18.

    Heinemann V, von Weikersthal LF, Decker T, et al. FOLFIRI plus cetuximab or bevacizumab for advanced colorectal cancer: final survival and per-protocol analysis of FIRE-3, a randomized clinical trial. Br J Cancer 2021;124:587–594.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19.

    Venook AP, Niedzwiecki D, Innocenti F, et al. Impact of primary (1°) tumor location on overall survival (OS) and progression-free survival (PFS) in patients (pts) with metastatic colorectal cancer (mCRC): analysis of CALGB/SWOG 80405 (Alliance) [abstract]. J Clin Oncol 2016;34(Suppl):Abstract 3504.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20.

    Venook AP, Ou FS, Lenz HJ, et al. Primary (1°) tumor location as an independent prognostic marker from molecular features for overall survival (OS) in patients (pts) with metastatic colorectal cancer (mCRC): analysis of CALGB/SWOG 80405 (Alliance) [abstract]. J Clin Oncol 2017;35(Suppl):Abstract 3503.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21.

    Grassadonia A, Di Marino P, Ficorella C, et al. Impact of primary tumor location in patients with RAS wild-type metastatic colon cancer treated with first-line chemotherapy plus anti-EGFR or anti-VEGF monoclonal antibodies: a retrospective multicenter study. J Cancer 2019;10: 5926–5934.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22.

    Fiala O, Ostasov P, Hosek P, et al. The predictive role of primary tumour sidedness in metastatic colorectal cancer treated with targeted agents. Anticancer Res 2019;39:5645–5652.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23.

    Lacouture ME, Mitchell EP, Piperdi B, et al. Skin toxicity evaluation protocol with panitumumab (STEPP), a phase II, open-label, randomized trial evaluating the impact of a pre-emptive skin treatment regimen on skin toxicities and quality of life in patients with metastatic colorectal cancer. J Clin Oncol 2010;28:1351–1357.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24.

    Guinney J, Dienstmann R, Wang X, et al. The consensus molecular subtypes of colorectal cancer. Nat Med 2015;21:1350–1356.

  • 25.

    Fontana E, Nyamundanda G, Cunningham D, et al. Molecular subtype assay to reveal anti-EGFR response sub-clones in colorectal cancer (CRC) [abstract]. J Clin Oncol 2018;36(Suppl):Abstract 658.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 26.

    Lenz HJ, Ou FS, Venook AP, et al. Impact of consensus molecular subtyping (CMS) on overall survival (OS) and progression free survival (PFS) in patients (pts) with metastatic colorectal cancer (mCRC): analysis of CALGB/SWOG 80405 (Alliance) [abstract]. J Clin Oncol 2017;35(Suppl):Abstract 3511.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27.

    Aderka D, Stintzing S, Heinemann V. Explaining the unexplainable: discrepancies in results from the CALGB/SWOG 80405 and FIRE-3 studies. Lancet Oncol 2019;20:e274–283.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28.

    Luhn P, Kuk D, Carrigan G, et al. Validation of diagnosis codes to identify side of colon in an electronic health record registry. BMC Med Res Methodol 2019;19:177.

    • Crossref
    • Search Google Scholar
    • Export Citation
Copyright:
Copyright © 2022 by the National Comprehensive Cancer Network 2022
Page Numbers:
8
Article Category:
Research Article
Received:
26 Mar 2021
Accepted:
01 Jul 2021
Print Publication Date:
01 Mar 2022
Online Publication Date:
04 Feb 2022
All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 3972 1205 20
PDF Downloads 1819 865 13
EPUB Downloads 0 0 0