番茄社区

Omitting bolus 5-fluorouracil (5-FU) from combination chemotherapy for gastrointestinal (GI) cancers did not adversely affect survival and was associated with less toxicity, a review of a large commercial database showed.

Across multiple types of GI cancers, patients who did not receive bolus 5-FU had overall survival that was almost identical to that of patients who did receive the bolus (HR 0.99, 95% CI 0.91-1.07). Leaving out the bolus was associated with significantly less neutropenia, thrombocytopenia, and use of growth factors.

The study confirms the current practice for some GI oncologists when using combination chemotherapy regimens, reported Shun Yu, MD, of NYU Langone Medical Center in New York City, and colleagues in the .

"I think only a minority of GI oncologists are doing this [omitting the bolus] right now," Yu told 番茄社区. "Another study that was done previously showed that oncologists who are more specialized in GI and oncologists who are more experienced to do this, but it's still a minority."

"Obviously, this is a retrospective study and we would need a large randomized controlled trial to validate or confirm some of these findings, but I think [our study] does confirm what a lot of us were already thinking," he added. "If you look at the history of how the bolus was incorporated and different pharmacokinetics, it really doesn't make sense. The half-life of 5-FU is pretty short, so the need for a bolus doesn't seem to be pharmacologically necessary. This gives us a little more evidence to pursue it in practice."

The study makes an important contribution for clinical practice, said Suneel D. Kamath, MD, of the Cleveland Clinic, who was not involved in the study.

"Most of us have suspected the 5-FU bolus doesn't add much efficacy and we know about the increased toxicity for bone marrow suppression and vomiting/diarrhea," said Kamath. "However, most oncologists continue to give it, probably because these cancers are very lethal and we don't want to be less 'aggressive.' This is especially true for those receiving adjuvant chemotherapy, where we are going for cure."

"These data go beyond just confirming current practice," he added. "The study shows that 86.3% of the 11,765 patients included did receive 5-FU bolus and only 13.7% did not. So clearly many oncologists who might think they can omit the 5-FU bolus are still giving it. Hopefully this study will give them the confidence to omit it more routinely."

Exclusion of patients receiving adjuvant 5-FU from the analysis leaves a major gap in knowledge.

"That is where the most trepidation about withholding bolus 5-FU lies and it would have been nice to see data showing survival is preserved there," said Kamath. "I'm sure this database has those patients in it, and I wonder why they didn't study that too and stratify by stage of disease and adjuvant versus palliative-intent chemotherapy."

Rationale for Bolus 5-FU

Recent advances in chemotherapy have not eliminated the need for the decades-old drug 5-FU. Multidrug regimens that include 5-FU remain "cornerstones" of treatment for GI cancers, Yu and co-authors noted in their introduction. National Comprehensive Cancer Network guidelines recommend 5-FU-containing regimens (such as , , and ) as first-line treatment for GI cancers.

The pharmacokinetic rationale for the 5-FU bolus is to minimize time to steady-state therapeutic drug concentration, the authors explained. However, 5-FU has a half-life of 8 to 20 minutes, and therapeutic levels occur quickly with infusional dosing. Mechanistically, the primary effect of the bolus is thought to be on RNA synthesis, whereas continuous infusion is thought to inhibit thymidylate synthesis. Whether both mechanisms are necessary has remained unclear.

Few studies have evaluated whether omission of the 5-FU bolus compromises efficacy of 5-FU-containing regimens. Prior retrospective studies showed conflicting results and were limited by small sample sizes and an inability to adjust for baseline differences between treatment arms.

To help inform decisions about a common clinical question, Yu and colleagues analyzed data from the database, which contains de-identified patient-level data from approximately 280 cancer clinics in the U.S. The analysis included adults with diagnoses of colorectal, pancreatic, or gastroesophageal cancers from 2011 to 2022. Only patients who received FOLFOX, FOLFIRI, or FOLFIRINOX were included.

Key Findings

Data analysis included 11,765 patients who had a median age of 63. Men accounted for 59.6% of the patients, and 64.2% were white. The cohort comprised 8,670 patients with colorectal cancer, 1,614 with pancreatic cancer, and 1,481 with gastroesophageal cancer.

FOLFOX was the most commonly used multidrug regimen (68.1%), followed by FOLFIRI (18.2%), and FOLFIRINOX (13.7%). The 5-FU bolus was included for 90.4% of patients who received FOLFOX, 89.2% of those who received FOLFIRI, and 61.6% of those who received FOLFIRINOX.

An unadjusted analysis showed no difference in OS between patients who received the 5-FU bolus and those who did not for colorectal cancer (23.6 vs 24.5 months), gastroesophageal cancer (10.4 vs 10.2 months), or pancreatic cancer (8.9 vs 9.5 months). For the entire cohort, patients who received the 5-FU bolus did have longer OS (20.3 vs 14.0 months; HR 0.74, 95% CI 0.69-0.79, P<0.01).

To balance covariates, the investigators generated a propensity score for all 11,745 patients. After balancing, OS among patients who received the 5-FU bolus was no longer improved (HR 0.99, 95% CI 0.91-1.07, P=0.74). The analysis also showed no significant difference in OS among prespecified subgroups, including those for age, sex, performance status, cancer type, multidrug regimen, and practice setting. Additional analyses showed no difference in OS by year of treatment or by the chemotherapy regimen.

The unadjusted analysis showed that patients who received the 5-FU bolus had a higher incidence of neutropenia within 14 days of treatment (22.3% vs 14.6%, P<0.01), including grade 3/4 neutropenia (8.2% vs 6.0%, P<0.01). Rates of anemia and thrombocytopenia did not differ significantly. The propensity-weighted analysis still showed a significantly higher rate of neutropenia among patients who received the bolus (22.7% vs 10.7%, P<0.01). Thrombocytopenia also occurred more often with the bolus (16.1% vs 11.2%, P<0.01), but anemia did not (80.2% vs 81.4%).

Given the concomitant impact of oxaliplatin and irinotecan on myelosuppression, the investigators performed an analysis adjusted for doses of the two drugs. The 5-FU bolus remained significantly associated with a higher risk of neutropenia (P=0.02) and thrombocytopenia (P<0.01).

Comment