Trends in Use and Perceptions About Triplet Chemotherapy Plus Bevacizumab for Metastatic Colorectal Cancer

Key Points Question What is the current rate of adoption of first-line systemic treatment with fluorouracil, folinic acid, oxaliplatin, and irinotecan plus bevacizumab (FOLFOXIRI-B) in daily practice? Findings In this cross-sectional study of 282 patients in the Netherlands, FOLFOXIRI-B prescription rates marginally increased in the past 5 years. During the study period, 1 in 7 estimated, eligible patients was treated with FOLFOXIRI-B. Meaning The results of this study suggest that, despite evidence of the effectiveness of FOLFOXIRI-B therapy, use of this therapy for metastatic colorectal cancer remains low, possibly because of oncologists’ reported preference for doublet therapy.


Introduction
In the past 2 decades, the choice of first-line treatment of patients with metastatic colorectal cancer (mCRC) has become more complex because of the increased availability of treatment options. 1  showing an improvement in progression-free survival of 2.6 months and, as a secondary end point, improved overall survival of 4 months. 2,3 Since publication of the TRIBE trial in 2014, the triplet regimen plus bevacizumab has been incorporated into national and international guidelines. 1,[5][6][7][8] Additional evidence for FOLFOXIRI-B was recently provided by the TRIBE2 (Upfront FOLFOXIRI Plus Bevacizumab and Reintroduction After Progression Versus mFOLFOX6 Plus Bevacizumab

Followed by FOLFIRI Plus Bevacizumab in the Treatment of Patients With Metastatic Colorectal
Cancer) trial in 2020, 4 which demonstrated that upfront FOLFOXIRI-B, followed by reintroduction after first disease progression, was superior to sequential administration of fluorouracil, oxaliplatin, and folinic acid plus bevacizumab (FOLFOX-B) and FOLFIRI-B (statistically significant benefit in progression-free survival 2 of 2.8 months and, as a secondary end point, in overall survival of 4.9 months). The efficacy of this regimen was confirmed in a meta-analysis 9 that included all available data from randomized clinical trials. This regimen was subsequently approved in 2020 by Dutch health care authorities. 10 No subgroup has been identified for which no benefit is expected, 9 although efficacy may be compromised in patients previously treated with oxaliplatin-based adjuvant chemotherapy. Because FOLFOXIRI-B has not been prospectively compared to doublet chemotherapy plus an antibody to the epidermal growth factor receptor in patients with left-sided primary and RAS (OMIM 164790, 190070, and 19002090070)/BRAF V600 (OMIM 164757) wild-type tumors, its benefit in this subgroup is uncertain.
Compared with a doublet chemotherapy backbone, FOLFOXIRI-B leads to a higher incidence of grades 3 to 4 diarrhea, (febrile) neutropenia, nausea, and mucositis. 9 Furthermore, FOLFOXIRI-B treatment involves the need for a central venous catheter and biweekly schedule. The latter is a relevant difference to the frequently used, 3-weekly capecitabine-based regimen. The published data justify FOLFOXIRI-B as a first-line treatment for eligible patients. However, the benefit-burden tradeoff presumably differs per individual.
In this study, we assess the adoption of FOLFOXIRI-B treatment for patients with mCRC in clinical practice in 3 parts. First, we assess the prescription of FOLFOXIRI-B from January 1, 2015, to December 31, 2018. Second, we explore the current prescription of FOLFOXIRI-B (November 1, 2020, to January 31, 2021). To provide a true representation of current practice, we used a cross-sectional

Study Design
The objective of this cross-sectional study was to assess the adoption of FOLFOXIRI Therefore, further mention of FOLFOXIRI-B in this article also includes FOLFOXIRI (eMethods in Supplement 1).

Interviews With Medical Oncologists
All medical oncologists who treat patients with mCRC in the Netherlands were invited by email or a direct colleague for an interview during the study week. Medical oncologists were initially not informed about the exact research question of this study to prevent information bias. They were simply informed that the interview would concern their personal approach to first-line systemic

Statistical Analysis
Data were exported from the electronic case report form to SPSS, version 27.0 (SPSS Inc). Descriptive statistics were used to present baseline characteristics. A 2-tailed, unpaired t test was used to compare continuous variables (age and FAS) between groups. A χ 2 test was used for group comparison of categorical variables. When groups were too small, we used the Fisher exact test. A 2-sided P < .05 was considered statistically significant. week. Baseline characteristics of medical oncologists are presented in Table 3.

Discussion of FOLFOXIRI-B With Patients
Fourteen medical oncologists (14%) reported that they did not discuss FOLFOXIRI-B with their patients.

Integration of Scientific Knowledge
To assess scientific knowledge integration, we asked questions regarding awareness of FOLFOXIRI-B as a treatment option (Table 4 and eTable 2 in Supplement 1). A total of 55 of 101 medical oncologists    patients, 56% also report that they regularly communicate a preference for a chemotherapy doublet to these patients. In addition, oncologists reported that they were more inclined to discuss FOLFOXIRI-B with patients with potentially resectable metastases, younger patients, and patients with more aggressive disease, such as patients with BRAF V600 -variant tumors. The previously mentioned reasons might explain the observed difference between the proportion of patients with whom FOLFOXIRI-B was discussed and the patients who actually received this treatment.

JAMA Network Open | Oncology
The assumption that FOLFOXIRI-B might be more efficacious for patients with BRAF V600variant tumors is based on the post hoc subgroup analysis of the updated TRIBE trial, which was subsequently incorporated in clinical guidelines. 2,3 However, a recent subgroup analysis 9 of individual patient data, which included 1697 patients, concluded that no increased benefit is     observed for patients with BRAF V600 -variant tumors. This meta-analysis 9 also assessed whether treatment benefit was larger for patients with liver-only metastasis. In addition, although R0 resections were more frequent (16.4% vs 11.8%), no significant interaction was found between treatment with FOLFOXIRI-B and the achievement of R0 resection in terms of overall survival.
When FOLFOXIRI-B is discussed as a treatment option, there are reasonable arguments as to why a patient might choose to forgo treatment with FOLFOXIRI-B. Medical oncologists reported increased toxic effects as the major reason for patients to decline FOLFOXIRI-B treatment, followed by the more frequent hospital visits from the 2-weekly schedule and the necessity of a central venous catheter. The discussion between oncologists and patients on the possible benefits and burden of a specific treatment can be challenging. When the condition of the patient allows several treatment options, personal beliefs, cultural aspects, and expectations of both physician and patient play a role.
The goal should be that information provided to patients is based on objective data to allow patients to make a choice that has the best chance of meeting personal goals.
Indeed, the adverse effects that are discussed by medical oncologists in our study are mostly in accordance with the TRIBE2 trial. 4 However, it is unknown how strongly toxic effects are emphasized by medical oncologists in their discussion with patients. In this respect, 44% of medical oncologists report to have initially overestimated the toxic effects for patients compared with 6% of medical oncologists who reported an initial underestimation when prescribing FOLFOXIRI-B. This finding suggests that Dutch medical oncologists might overemphasize toxic effects in daily practice.
Another factor related to the relatively low adoption rate might be the transfer of scientific advancement to daily practice. Typically, knowledge is generated through clinical studies. Published results are subsequently included in the synthesized literature base and then broadly disseminated through conferences, workshops, and continuing medical education. Finally, when incorporated as standard of care, said insights may ultimately lead to improved health outcomes. Several estimates have suggested that this entire process takes up to 17 years to complete. 23-26 Considering the current rate of the development of novel and effective cancer treatments, the adoption of these treatments in daily practice should be much improved. Our data indicate that medical oncologists gained knowledge on the benefit of FOLFOXIRI-B through a various number of sources, if at all. We found that medical oncologists who were aware of more literature on FOLFOXIRI-B seemed to be less inclined to communicate a preference for treatment with chemotherapy doublets to a patient, resulting in patients following that recommendation. This finding could imply that a higher level of awareness of medical oncologists leads to a more unbiased discussion of treatment options with patients.
Continuing medical education, feedback on guideline adherence, modular updates of guidelines, and more frequent dissemination of novel, positive developments through authorized organizations should aid in this process. These types of interventions are known to correspond with guideline dissemination and implementation with significant positive changes in patient outcomes. 27 In addition, real-life data should provide up-to-date information on the implementation of novel, effective treatments in daily practice and will provide more transparency on the quality of cancer care in individual hospitals. Flash mob studies, such as the current study, may be useful in achieving this goal.
This study is, to our knowledge, the sixth and largest flash mob study ever performed worldwide. [13][14][15][16][17] Ultrafast data generation coupled with thoroughly preplanned data analysis enabled us to provide a true snapshot of current medical practice.