SUMMARY: ColoRectal Cancer (CRC) is the third most common cancer diagnosed in both men and women in the United States. The American Cancer Society estimates that approximately 154,270 new cases of CRC will be diagnosed in the United States in 2025 and about 52,900 patients will die of the disease. The lifetime risk of developing CRC is about 1 in 23. Among patients with Stage II-III CRC, 20-40% will develop metastatic disease.
Approximately 80% of patients present with localized disease, offering an opportunity for curative-intent therapy through surgical resection followed by adjuvant or neoadjuvant treatment approaches. The current standard of care for localized colon cancer involves surgery followed by risk-adapted adjuvant chemotherapy, while patients with localized rectal cancer may receive neoadjuvant chemoradiotherapy or chemotherapy before surgical resection. In select cases, a nonoperative “watch-and-wait” strategy is pursued to preserve organ function following a complete clinical response to neoadjuvant therapy.
Among available chemotherapy regimens, Capecitabine plus Oxaliplatin (CAPOX) has become a mainstay for patients with localized CRC in both adjuvant and neoadjuvant settings.
Rationale for CAPOX Use
Capecitabine (XELODA®), an oral fluoropyrimidine prodrug, is metabolized to 5-Fluorouracil (5-FU) in the liver and tumor tissue. Oxaliplatin, a platinum analog, induces cytotoxicity through DNA crosslinking, leading to apoptosis. The combination of these agents results in synergistic antitumor activity and has demonstrated robust efficacy in CRC.
The pivotal International Duration Evaluation of Adjuvant Therapy (IDEA) collaboration established that 3 months of CAPOX provides similar efficacy to 6 months of Oxaliplatin-based therapy in the low-risk subgroup of patients, while significantly reducing cumulative neurotoxicity, treatment burden, and cost. These findings drove widespread adoption of CAPOX as the preferred regimen for adjuvant treatment of localized CRC. Furthermore, studies such as RAPIDO and OPRA have supported CAPOX use in the neoadjuvant management of rectal cancer.
Despite this, real-world tolerability data from U.S. patients remain limited, and prior evidence suggests that Capecitabine-based regimens may be less well tolerated in American populations, particularly among women. These gaps highlight the need to better understand how CAPOX performs in U.S. clinical practice.
Study Objective
This retrospective, single-institution analysis sought to evaluate the real-world tolerability of CAPOX in patients with localized CRC treated with curative intent, either in the adjuvant or neoadjuvant setting.
Methods
- Design and Setting:
Retrospective cohort study including patients treated across 17 academic and community oncology clinics within a single cancer care network. - Eligibility Criteria:
Adults (≥18 years) with Stage II or III colon or rectal cancer who initiated CAPOX between June 1, 2017, and June 1, 2023, and received at least one treatment cycle. - Endpoints:
- Primary endpoint: Completion of all intended CAPOX cycles (as determined by treating clinicians, regardless of dose modification).
- Secondary endpoints: Incidence of grade ≥3 adverse events, hospitalizations related to treatment toxicity, and dose reductions.
- Data Collection:
Patient demographics, tumor characteristics, treatment details, and toxicity data were extracted from electronic medical records and pharmacy databases. Creatinine clearance was calculated via the Cockcroft-Gault formula to guide capecitabine dose adjustments.
Patient Population
A total of 153 patients were included (median age: 61 years).
- Sex: 51% male, 49% female
- Race: 81% White, 15.7% Black
- Tumor site: 63% colon, 37% rectal
- Stage: 21.6% Stage II, 78.4% Stage III
- ECOG performance status: 0 (42.5%), 1 (50%)
- Renal function: Majority with creatinine clearance ≥50 mL/min; 3.9% had CrCl 30–50 mL/min and received upfront dose adjustment.
Key Findings
- Treatment Completion:
- Only 44.4% (95% CI, 36–52) completed all intended CAPOX cycles.
- Completion was lower among female patients (34.6%) compared with the overall cohort.
- Completion rates varied by treatment duration:
- 4 cycles: 55% (95% CI, 43–66)
- 6 cycles: 41% (95% CI, 23–59)
- 8 cycles: 33% (95% CI, 20–45)
- Dose Modifications:
- 24% received upfront dose adjustments for renal function or performance status.
- Average starting doses of both agents were comparable between patients who completed versus discontinued treatment.
- Toxicity:
- Grade ≥3 adverse events: 30.7% (95% CI, 23–38)
- Hospitalizations due to toxicity: 17.6% (95% CI, 11–23)
- Premature discontinuation: 21.5% stopped Oxaliplatin early; 40.5% discontinued both agents.
- Regimen modification: 6.5% switched to FOLFOX or 5-FU/Leucovorin; 27% continued single-agent Capecitabine.
- Predictors of Completion:
Multivariable analysis identified race, sex, and intended cycle number as independent predictors of treatment completion.
Interpretation and Context
The completion rate in this real-world study (44%) was markedly lower than the ~85% and 64% completion rates observed in the 3- and 6-month CAPOX arms, respectively, of the IDEA trial. Differences in patient selection, clinical setting, and regional tolerability may explain this discrepancy.
Notably, female sex was associated with lower tolerability, echoing prior reports that women experience greater Fluoropyrimidine-related toxicity. This may relate to lower lean body mass relative to body surface area, resulting in higher effective drug exposure, as well as potential differences in drug metabolism, hormonal influence, and cultural reporting patterns.
Racial differences were also observed, with White patients demonstrating poorer tolerability compared with Black patients, findings that contrast with some previous single-center U.S. studies and warrant further exploration.
Interestingly, age ≥70 years and ECOG performance status ≥1 were not significantly associated with early discontinuation, although the small number of elderly patients limits conclusions.
Clinical Relevance
These findings underscore that toxicity remains a major barrier to completing CAPOX therapy in U.S. clinical practice. Given the importance of treatment dose intensity for cure, optimizing patient selection and providing proactive supportive care are crucial to maximizing therapeutic benefit.
The data also call for reconsideration of CAPOX versus FOLFOX selection in certain patient subsets. While CAPOX offers the convenience of oral administration and shorter duration, FOLFOX may be more tolerable in select populations, particularly women or patients at risk for Capecitabine toxicity.
Limitations
This study was retrospective and single-institution in nature, with potential for incomplete data capture and limited generalizability. Lack of DPYD genotype testing prevented assessment of pharmacogenetic contributions to toxicity. Despite these limitations, inclusion of both academic and community centers enhances the representativeness of findings.
Conclusion
In this large, real-world analysis, fewer than half of patients with localized CRC receiving CAPOX completed their intended cycles, primarily due to treatment-related toxicity. Female sex, White race, and longer planned therapy duration were associated with lower completion rates. These data highlight the need for personalized treatment strategies, vigilant toxicity monitoring, and robust supportive measures to ensure optimal delivery of curative-intent chemotherapy.
As CAPOX continues to anchor adjuvant and neoadjuvant treatment protocols for CRC, understanding its real-world tolerability will remain essential for aligning evidence-based guidelines with practical patient care realities in oncology practice.
Real-World Tolerability of Capecitabine and Oxaliplatin in Patients in the United States With Localized Colorectal Cancer Undergoing Curative-Intent Treatment. Mears V, Naleid N, Pawar O, et al. JCO Oncol Pract 2025;21:1355-1363
