[Skip to Navigation]
Sign In
Invited Commentary
February 14, 2020

Bringing Personalized Medicine to Precision Medicine in Gastroesophageal Cancer

Author Affiliations
  • 1Department of Medical Oncology and Therapeutics Research, City of Hope Comprehensive Cancer Center, Duarte, California
  • 2Department of Medicine, Massachusetts General Hospital Cancer Center, Boston
  • 3Harvard Medical School, Boston, Massachusetts
JAMA Netw Open. 2020;3(2):e1921289. doi:10.1001/jamanetworkopen.2019.21289

Perioperative chemotherapy remains critical in improving long-term disease-free and overall survival outcomes among patients with locally advanced yet surgically resectable gastroesophageal adenocarcinoma (GEA). Fluorouracil, leucovorin, oxaliplatin, and docetaxel (FLOT) is a recognized standard treatment from the phase 3 FLOT4 trial,1 but toxic effects limit its application to select patients such that doublet chemotherapy with fluorouracil, leucovorin, and oxaliplatin (FOLFOX) is also recommended in national guidelines.2 Catenacci et al3 present a single-group phase 2 trial exploring an alternative triplet chemotherapy regimen of fluorouracil, leucovorin, oxaliplatin, and irinotecan (FOLFIRINOX), in which all 3 cytotoxic agents have demonstrable activity in the metastatic setting. Thus, the report by Catenacci et al3 presents a unique first application of this combination among patients with GEA that can be surgically resected with curative intent. Furthermore, the authors pursued a personalized pharmacogenomic approach in dosing the irinotecan (gFOLFIRINOX) based on the UGT1A1 genotype (OMIM 191740), using standardized assays for detection of germline polymorphisms in the promoter region or first coding exon of UGT1A1. Common polymorphisms include variations in the number of TA repeats in the promoter region, which influence the activity of the UGT1A1 enzyme (ie, genotype 6/6 harbors 6 repeats and has normal enzyme activity; genotype 7/7 equates to 7 repeats and reduced activity), which is relevant to the elimination of the active metabolite of irinotecan, SN-38. Patients with homozygous germline polymorphisms (ie, genotype 7/7) exhibit a more severe phenotype, susceptible to increased toxic effects from irinotecan. Building on their work in the metastatic setting, Catenacci et al3 pursued an up-front dose reduction of irinotecan for patients with genotyping results associated with diminished activity of the UGT1A1 enzyme. In enrolling patients at 2 US centers, it is not surprising that 31 of 38 patients were white individuals, with 3 patients exhibiting the higher risk of toxic effects of the 7/7 genotype. Only 2 patients were of Asian descent, with 1 patient in the 7/7 genotype group. Although the known polymorphism in exon 1 of the UGT1A1 gene, which is also associated with diminished enzyme activity, was included in their assay platform, it was not found in any patients enrolled in the trial. However, the pharmacogenomic dosing used in this study can still serve as a framework in Asian countries, where GEA has higher incidence and both TA promoter repeats and exon 1 coding polymorphisms encompass the differing UGT1A1 genotypes.

Regarding their observed efficacy outcomes, Catenacci et al3 used the feasible and clinically relevant coprimary end points of margin-negative resection rate and pathologic response grade (PRG, also commonly referred to as tumor regression grade [TRG]) to power their single-group statistical analysis of 36 evaluable patients. For their intention-to-treat population, 33 of 36 patients (92%) met the investigators’ goal of at least 30 patients achieving a margin-negative resection, although 3 patients (8%) achieving a complete pathologic response (ie, PRG 1a) fell short of their target of 4 patients to reject the null hypothesis. While this study had a small sample, it is interesting to draw comparisons with the FLOT4 trial data set,1 which reported that 6% of patients achieved PRG 1a with the triplet of epirubicin, cisplatin, and fluorouracil (ECF) and 16% achieved PRG 1a with FLOT. With median follow up being relatively early (ie, approximately 20 months), median disease-free survival appeared promising at 30.1 months and median overall survival had yet to be reached. For the adjuvant phase of therapy, 5 patients (14%) were prescribed regimens other than gFOLFIRINOX, including 4 patients (11%) who received FLOT, which may skew interpretation of long-term survival outcomes. However, the consequences of the change in adjuvant therapy may be minimal, given that, across the spectrum of perioperative trials, dose intensity is 80% to 90% in neoadjuvant courses of therapy versus 50% to 60% in adjuvant courses of therapy, leading some to argue that survival benefits are derived predominantly from the neoadjuvant phases of therapy.4 Genotype-based selection of FOLFIRINOX dosing encouragingly did not impact efficacy outcomes. However, among the 3 patients with genotype 7/7, a 67% rate (ie, 2 of 3 patients) of grade 3 or higher diarrhea was observed, even with the up-front dose reduction of irinotecan, in contrast to a 16% and 13% rate of this same toxic effect in the patients with genotypes 6/6 and 6/7, respectively. With FLOT, oncologists have had to acquaint themselves with properly managing diarrhea because grade 3 or 4 diarrhea significantly increased in incidence in the FLOT arm (10%) of the FLOT4 trial compared with the control arm of ECF (4%).1 As such, knowledge of a patient’s UGT1A1 genotype appears paramount before proceeding with prescribing FOLFIRINOX in this setting.

Catenacci et al3 also explored incorporation of the anti-ERBB2 (formerly HER2) monoclonal antibody trastuzumab in 6 patients, who had tumors with ERBB2 overexpression, given the known survival advantage with the addition of anti-ERBB2 therapy to first-line chemotherapy in metastatic patients whose cancers harbor the ERBB2 biomarker. Intriguingly, PRG rates were not uniform in this group, with only 1 patient (17%) exhibiting a complete pathologic response, which contrasts with the high rates (ie, 40%) observed when trastuzumab is added to neoadjuvant chemotherapy in breast cancer.5 However, higher intratumoral heterogeneity of ERBB2 overexpression, among other oncogenic drivers, is becoming increasingly recognized as accounting for the variable success of molecularly targeted strategies in gastroesophageal versus breast cancer in the metastatic setting.6 Larger data sets regarding the incorporation of anti-ERBB2 strategies in perioperative therapy for GEA are yet to mature, but if suboptimal outcomes with molecularly targeted approaches are also observed in the nonmetastatic setting, a query into the effects of intratumoral heterogeneity is likely warranted. Data sets have already emerged that show that, even in early-stage nonmetastatic de novo GEA presentation, a high degree of intratumoral heterogeneity of relevant oncogenic drivers already preexists.7

Moving forward, for patients eligible for more intense 3-drug vs 2-drug perioperative chemotherapy, it is difficult to envision that FOLFIRINOX will handily supplant FLOT in treatment guidelines based on this data set alone. However, among patients with contraindications to taxanes who are still being considered for more aggressive 3-drug therapy, gFOLFIRINOX may be a suitable selection for those without low-metabolizing UGT1A1 genotypes. Larger prospective randomized clinical trials will need to be considered to address whether gFOLFIRINOX can be considered more advantageous than FOLFOX in the perioperative setting. Similarly, ongoing total neoadjuvant therapy (TNT) trials in patients with GEA may yet improve on current outcomes and nullify the concerns over the tolerability of the adjuvant component of perioperative approaches. In the era of precision medicine, efforts to purely optimize preexisting cytotoxic drug choices (ie, gFOLFIRINOX vs FLOT vs FOLFOX) may not be the best use of patient and health care resources. The ideal future state will be the application of multiomic patient and tumor testing to pair personalized drug dosing with precision targeting of the disease drivers to achieve equipoise between maximizing therapeutic benefit and minimizing toxic effects. This study of genotype-based dosing carried out by Catenacci et al4 provides a helpful paradigm to frame future research efforts.

Back to top
Article Information

Published: February 14, 2020. doi:10.1001/jamanetworkopen.2019.21289

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Chao J et al. JAMA Network Open.

Corresponding Author: Joseph Chao, MD, Department of Medical Oncology and Therapeutics Research, City of Hope Comprehensive Cancer Center, 1500 E Duarte Rd, Duarte, CA 91010 (jchao@coh.org).

Conflict of Interest Disclosures: Dr Chao reported receiving personal fees and grants from Merck and Co and receiving personal fees from Daiichi Sankyo, Taiho Pharmaceutical, and Eli Lilly and Co outside the submitted work. Dr Klempner reported receiving personal fees and serving on the advisory boards of Eli Lilly and Co, Astellas, Foundation Medicine, and Bristol-Myers Squibb; receiving personal fees and consulting fees and serving on the advisory board of Boston Biomedical; and owning stock in Turning Point Therapeutics. No other disclosures were reported.

Al-Batran  SE, Homann  N, Pauligk  C,  et al; FLOT4-AIO Investigators.  Perioperative chemotherapy with fluorouracil plus leucovorin, oxaliplatin, and docetaxel versus fluorouracil or capecitabine plus cisplatin and epirubicin for locally advanced, resectable gastric or gastro-oesophageal junction adenocarcinoma (FLOT4): a randomised, phase 2/3 trial.  Lancet. 2019;393(10184):1948-1957. doi:10.1016/S0140-6736(18)32557-1PubMedGoogle ScholarCrossref
Ajani  JA, D’Amico  TA, Bentrem  DJ,  et al.  Esophageal and esophagogastric junction cancers, version 2.2019, NCCN Clinical Practice Guidelines in Oncology.  J Natl Compr Canc Netw. 2019;17(7):855-883. doi:10.6004/jnccn.2019.0033PubMedGoogle ScholarCrossref
Catenacci  DVT, Chase  L, Lomnicki  S,  et al.  Evaluation of the association of perioperative UGT1A1 genotype–dosed gFOLFIRINOX with margin-negative resection rates and pathologic response grades among patients with locally advanced gastroesophageal adenocarcinoma: a phase 2 clinical trial.  JAMA Netw Open. 2020;3(2):e1921290. doi:10.1001/jamanetworkopen.2019.21290Google Scholar
Chuang  J, Gong  J, Klempner  SJ, Woo  Y, Chao  J.  Refining the management of resectable esophagogastric cancer: FLOT4, CRITICS, OE05, MAGIC-B and the promise of molecular classification.  J Gastrointest Oncol. 2018;9(3):560-572. doi:10.21037/jgo.2018.03.01PubMedGoogle ScholarCrossref
Cortazar  P, Zhang  L, Untch  M,  et al.  Pathological complete response and long-term clinical benefit in breast cancer: the CTNeoBC pooled analysis.  Lancet. 2014;384(9938):164-172. doi:10.1016/S0140-6736(13)62422-8PubMedGoogle ScholarCrossref
Zhao  D, Klempner  SJ, Chao  J.  Progress and challenges in HER2-positive gastroesophageal adenocarcinoma.  J Hematol Oncol. 2019;12(1):50. doi:10.1186/s13045-019-0737-2PubMedGoogle ScholarCrossref
Pectasides  E, Stachler  MD, Derks  S,  et al.  Genomic heterogeneity as a barrier to precision medicine in gastroesophageal adenocarcinoma.  Cancer Discov. 2018;8(1):37-48. doi:10.1158/2159-8290.CD-17-0395PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words