Should we still be using bolus 5-FU prior to infusional regimens in gastrointestinal cancers? A practical review

Abstract

5-fluorouracil (5-FU), a pyrimidine analogue with antimetabolite activity, is one of the most widely used drugs in Oncology and many different regimens have been described regarding its use. Nowadays, the modified de Gramont is the most popular schedule of 5-FU to treat gastrointestinal cancers and may be given either alone or combined with irinotecan, oxaliplatin and monoclonal antibodies. The true clinical value of bolus 5-FU right before infusional regimens remains to be determined since no randomized trials have addressed this issue. This manuscript aims to review the history of 5-FU, its mechanism of action and the data exploring the role of bolus 5-FU.

Introduction

5-fluorouracil (5-FU), a pyrimidine analogue with antimetabolite activity, is one of the most widely used drugs in Oncology [1]. First scientifically reported by Heidelberger in 1957 [2], when 5-FU demonstrated clinical activity, it has since been the treatment backbone of most gastrointestinal malignancies, both in curative-intent and palliative settings. Other primary tumors may also benefit from this class of drug, such as breast and head and neck cancers.

Although Moertel et al. had already reported in 1972 their double-blind randomized trial demonstrating a higher efficacy of a slower infusion of 5-FU (in 2 h) versus the bolus regimen among 149 colon cancer patients [3], the standard 5-FU administration schedule in the 1970s was either a weekly bolus or a 5-day bolus regimen at usually 4–5-week intervals. At that time, the rapid injection method was preferred for practical reasons. Indeed, the first approval scheme for 5-FU by FDA was an administration of 15 mg/kg × 4, then 7.5 mg/kg on alternate days according to a phase I study from 1958 [4]. Nonetheless, a mortality rate of almost 3% from 5-FU toxicity had been experienced with this FDA-approved schedule [5], leading to more research aiming at finding a better tolerated scheme.

Reports with a longer 5-day continuous infusion of 5-FU became available since the 1970s with the famous Wayne State University chemoradiation regimen including also mitomycin-C for anal cancer [6]. Such initial results boosted the conduct of a prospective trial by Seifert et al. in which 70 colorectal cancer patients received either an intravenous bolus injection of 5-FU daily for 5 days or a continuous 120-h infusion [7]. The prolonged infusion demonstrated higher response rates and less myelotoxicity, but required hospitalization, leading to higher costs and more inconvenience at that time.

In the 1980s, driven by increasing availability of central venous access devices and ambulatory infusion pumps for outpatient administration, infusional administration of 5-FU was more frequently prescribed. By that time and later on, more data on better outcomes with infusional regimens when compared to bolus regimens became available, both with randomized trials and meta-analyses [8–11]. Overall, response rates and overall survival were higher with infusional administration of 5-FU when compared to bolus schemes in colorectal cancer [8–11], reinforcing the notion that protracted infusions of 5-FU increase the duration of tumor cell exposure leading to maximal potential of tumor cell killing.

Another study compared the therapeutic ratio of a monthly schedule of low-dose leucovorin (LV) and 5-FU bolus with a bimonthly schedule of high-dose LV and 5-FU bolus plus continuous infusion in patients with advanced colorectal cancer. The results showed that the bimonthly regimen was more effective and less toxic than the monthly regimen, while it also increased the therapeutic ratio. However, there was no evidence of increased survival [12]. Those better clinical outcomes reported with longer infusions of 5-FU were in line with the concept of dose intensity, first popularized by Hryniuk in breast cancer studies, in which the critical determinant of response of 5-FU is the dose delivered over unit time [13]. However, it remains unknown whether duration of exposure to 5-FU is comparable to dose intensity [14].

Nowadays, the modified de Gramont regimen is the most popular schedule of 5-FU to treat gastrointestinal cancers and may be given either alone or combined with irinotecan (FOLFIRI), oxaliplatin (FOLFOX), both (FOLFOXIRI and FOLFIRINOX) or monoclonal antibodies. It involves a 2-h infusion of LV (400 mg/m2) followed by a push administration of 5-FU (400 mg/m2) on day 1 and then continuous 46-h infusion of 5-FU (2.400 mg/m2) administered every 14 days. The folinic acid, also known as LV, has demonstrated that its combination to 5-FU increase response rates and overall survival [15]. This drug, therefore, is commonly administered in several 5-FU protocols.

Mechanism of action of 5-FU

In a simplified way, anti-tumor activity of 5-FU depends on inhibition of both DNA synthesis and RNA function, while the balance of these effects relies on the modalities of administration. Other mechanisms play important roles in 5-FU activity, such as enzymes of 5-FU metabolism (e.g. dihydropyrimidine dehydrogenase; DPD, rate-limiting enzyme of catabolism) and action target (e.g. thymidylate synthase; TS) [1, 16]. Actually, TS inhibition has been considered the principal mechanism of 5-FU action. The 5-FU metabolite fluorodeoxyuridine monophosphate (FdUMP) competes with the pyrimidine uracil to form a stable complex with TS and the folate cofactor, leading to decreased thymidine production. Consequently, decreased DNA synthesis and repair occurs, reducing cell proliferation [17, 18]. LV (formyltetrahydrofolate) enhances activity of 5-FU by stabilizing the binding of FdUMP to TS. In other words, LV slows the catabolism of 5-FU via stabilization of the TS-FU-FdUMP complex with the result that the intracellular activity of FU is prolonged [19]. This causes a decrease in intracellular levels of that enzyme and a resulting drop in the production of thymidylate. In this way, LV can enhance or modulate the activity of fluorouracil [20]. Indeed, LV is important to increase the half-life of bolus 5-FU, although it does not have a role if infusional 5-FU alone is being used.

By simulating a uracil molecule with a fluorine atom at the C-5 position instead of hydrogen, 5-FU is incorporated into RNA and DNA, leading to cytotoxicity [19]. The other 5-FU metabolite, fluorodeoxyuridine triphosphate (FdUTP), is incorporated into DNA thus interfering with DNA replication. In parallel, fluorouridine-5-triphosphate (FUTP), another 5-FU metabolite, is incorporated into RNA in place of uridine triphosphate (UTP), producing a fraudulent RNA and interfering with RNA processing and protein synthesis [17].

DPD-mediated conversion of 5-FU to dihydrofluorouracil (DHFU), an inactive metabolite, is the rate-limiting step of 5-FU catabolism in both normal and tumor cells. Almost 80% of administered 5-FU is broken down by DPD in the liver [21]. Variations in DPD activity result in differences in toxicity.

While bolus 5-FU acts mainly by incorporation into RNA, infusional 5-FU causes more inhibition of TS. This may also explain the different 5-FU-related toxicities produced by each schedule, since bolus 5-FU tends to cause more leukopenia, mucositis, and diarrhea, while continuous infusion 5-FU is more associated with hand-foot syndrome [22]. In vitro models with flow cytometric DNA analysis have demonstrated that short-term followed by continuous exposure to 5-FU produce a marked increase in percentage of S-phase cells, compared with the percentage for each schedule alone, suggesting a synergistic effect [23]. Nonetheless, the true clinical value of bolus 5-FU right before infusional regimens remains to be determined.

Clinical data

In colorectal cancer, a variety of dosage schedules of FOLFOX have been used over time, with or without a monoclonal antibody. Nowadays, the mFOLFOX6, which consists of biweekly oxaliplatin (85 mg/m²), LV (400 mg/m²), and 5-FU (400 mg/m² administered by intravenous bolus, followed by 2400 mg/m² given as a 46-h continuous infusion), is the most prescribed regimen. A Japanese study aimed to compare mFOLFOX6 + bevacizumab with mFOLFOX7 (no bolus of 5-FU) + bevacizumab in 39 patients with metastatic colorectal cancer to assess the clinical significance of omitting bolus 5-FU [24]. With a better toxicity profile and similar outcomes, the authors raised the question whether bolus 5-FU could be definitely omitted in FOLFOX regimens.

Although the omission of 5-FU has not been directly tested against its use, the mFOLFOX7 (without 5-FU bolus) has already been part of phase III trials arms. In the CONcePT trial, which evaluated the role of intermittent oxaliplatin, both arms included mFOLFOX7 with bevacizumab [23]. In this study, the omission of bolus 5-FU resulted in a low incidence of severe myelosuppression (2%) [25].

More recently, a retrospective study compared clinical outcomes of 133 patients with metastatic colorectal cancer who received either mFOLFOX6 with bolus 5-FU and LV or the same regimen without both 5-FU bolus and LV. No differences were observed in progression free survival (PFS) or overall survival (OS) between the arms. Despite reduced growth factor utilization, the non-bolus group demonstrated a favorable safety profile with less treatment-related hematologic events [26]. Another recent retrospective study including 252 patients with metastatic colorectal cancer (who received mFOLFOX6 with or without bolus 5-FU) showed no impact of 5-FU bolus omission on PFS, although it did reveal a worse OS. Given the retrospective nature of the study and the fact that oncologists typically omit 5-FU bolus in older and more frail patients, results might have been influenced by those variables [27].

In metastatic pancreatic cancer, FOLFIRINOX has become the most preferred first line regimen for young and fit patients, and it consists of biweekly oxaliplatin (85 mg/m²), irinotecan (180 mg/m²), LV (400 mg/m²), and 5-FU (400 mg/m² administered by intravenous bolus, followed by 2400 mg/m² given as a 46-h continuous infusion) [28]. However, frequent adverse events with the classic FOLFIRINOX regimen have led many oncologists to omit 5-FU bolus and dose-reduce both irinotecan and oxaliplatin. A retrospective study evaluated a modified FOLFIRINOX schedule without the bolus of 5-FU, and administration of hematopoietic growth factors in 60 pancreatic cancer patients, 60% of whom metastatic. Besides showing a better safety profile, this modified FOLFIRINOX maintained efficacy, with an overall response rate of 30% and a median OS of 16.4 months [29]. Other modified FOLFIRINOX regimens without bolus 5-FU have been tested in the metastatic scenario with improved safety profile, including prospective phase II trials, and no signs of decreased activity have been found^, [29–31]. In the adjuvant setting, a modified FOLFIRINOX without 5-FU bolus and dose-reduced irinotecan (150 mg/m²) has become the most active regimen to decrease recurrence and increase survival following pancreatic adenocarcinoma resection [32]. In this important phase III trial, LV was maintained in the FOLFIRINOX regimen.

Conclusion

5-FU bolus and LV, when combined to infusional 5-FU regimens, do not seem to add activity, while it increases toxicity. Since randomized data are still lacking regarding the role of 5-FU bolus and LV, international efforts should be made to implement randomized trials investigating this matter.

Funding

No funding sources.

Declarations

Conflict of interest

No conflict of interests.