Retrospective Analysis of the Safety of FOLFOX Compared to CAPOX for Adjuvant Treatment of Stage III Colorectal Cancer in Newfoundland Patients

Abstract

Background: Capecitabine and oxaliplatin (CAPOX) and infusional 5-fluouracil, folinic acid, and oxaliplatin (FOLFOX) are the two chemotherapy regimens in current clinical use for the adjuvant treatment of colorectal cancer (CRC). Many centers in Newfoundland lack the resources to support the home infusion program required for FOLFOX, leaving CAPOX as the sole treatment option. This study aimed to review if Newfoundland patients receiving CAPOX experience greater treatment-induced toxicities. Methods: A multicenter retrospective cohort study of 93 Stage III CRC patients. The frequency and severity of toxicities, healthcare resource utilization, and treatment completion rates were compared between the two treatment options. Results: Grade 3 diarrhea and grade 1 or 2 nausea/vomiting were more common in CAPOX compared to FOLFOX-treated patients (26.9% versus 2.99%, p = 0.002; 61.5% versus 31.8%; p = 0.048, respectively). Grade 1 or 2 mucositis was more common with FOLFOX (35.8% versus 3.9%, p = 0.002). CAPOX was associated with higher rates of severe toxicity (53.9% versus 25.4%, p = 0.009), while rates of grade 1 and 2 toxicities were not significantly different between groups. CAPOX-treated patients were greater than twice as likely to require emergency department treatment secondary to toxicity (mean 0.692 visits per patient versus 0.313 in FOLFOX patients, p < 0.001) and the proportion of patients that were hospitalized secondary to CAPOX toxicity was greater. Significantly more FOLFOX patients were able to finish their prescribed treatment plans compared to CAPOX patients (89.5% versus 53.8%; p < 0.001). Conclusions: Compared to FOLFOX-treated patients, CAPOX patients are more likely to experience toxicities of greater severity, require emergency services secondary to treatment-related toxicity, and to discontinue therapy. This reflects a reduced standard of care that may decrease patient safety and quality of life.

1. Introduction

Colorectal cancer (CRC) is the third most commonly diagnosed cancer in Canada [1]. It is estimated that during 2021 24,800 new cases of CRC were diagnosed and more than 9600 Canadians died from the disease. Currently, surgery is the primary treatment for most early stage CRCs and although resection of Stage III CRC has curative intent, many patients experience disease relapse [2]. This recurrence is thought to arise from small, undetectable cancerous cells referred to as micrometastases. To reduce the risk of relapse, adjuvant chemotherapy regimens aimed at eradicating remaining cancerous cells have been adopted as the standard of care for most Stage III CRC patients. To date, the efficacy of adjuvant chemotherapeutics for Stage III CRC has been most well demonstrated by combining oxaliplatin with fluoropyrimidines [3,4,5] and two such treatment regimens, FOLFOX and CAPOX, are in current clinical use.

The FOLFOX regimen consists of intravenous (I.V) administration of oxaliplatin, leucovorin, and 5-fluorouracil (5-FU) with an additional 46 h ambulatory infusion of 5-FU via central line [6]. Its use for adjuvant Stage III CRC was approved in Canada following the results of the Multicenter International Study of Oxaliplatin/5-Fluorouracil/Leucovorin in the Adjuvant Treatment of Colon Cancer (MOSAIC) [3,6]. In contrast to FOLFOX, the CAPOX chemotherapy regime does not administer 5-FU I.V. CAPOX contains I.V oxaliplatin combined with an oral chemotherapy agent, capecitabine, a prodrug that is metabolized to 5-FU within the body.

Previous cancer care guidelines have supported six months of FOLFOX for adjuvant Stage III CRC as the standard of care with CAPOX being used primarily when resources or patient preference impedes the use of ambulatory infusion. This has been important in Newfoundland because large areas across the island lack the nursing resources to support a home infusion program, leaving CAPOX as the only option for therapy. However, in March 2018, data published in the International Duration Evaluation of Adjuvant Therapy (IDEA) trial suggested that in low-risk patients, three months of CAPOX chemotherapy was non-inferior to six months of FOLFOX chemotherapy [7]. Therefore, CAPOX has become more frequently used for many low-risk Stage III CRC patients.

5-FU with leucovorin and capecitabine has been shown to be equally effective with similar survival outcomes for adjuvant Stage III CRC [8], however their associated toxicities have never been compared by a direct head-to-head clinical trial. Previous retrospective reviews that have compared the toxicities of FOLFOX and CAPOX have demonstrated conflicting results with some failing to stratify the severity of chemotherapy induced toxicities [9,10] and others falling short of correlating these results to real world implications on treatment completion, patient safety, and healthcare resource utilization. Moreover, none of the existing studies incorporate the recommendations provided by the IDEA trial. As a result, their findings are based solely on comparing patients that received 6 months of either chemotherapy regimen. It is unknown what implication the current recommendation to reduce the duration of CAPOX therapy from six to three months in low-risk patients will have on toxicity when compared to six months of FOLFOX therapy.

The lack of a consensus and comprehensive comparison of the safety of FOLFOX compared to CAPOX has motivated this retrospective study. In this paper we seek to determine: Are the toxicities experienced by Stage III CRC patients receiving FOLFOX and CAPOX equivalent? If not, what implication do these treatment toxicities have on treatment delay, dose reduction, and completion of the prescribed chemotherapy regimen?

2. Results

Ninety-three Stage III CRC patients were included in this study. In total, 67 of these patients were prescribed the FOLFOX regimen while 26 received CAPOX (

Table 1. Patient Demographics and Comorbidities.

	FOLFOX	CAPOX	p Value
Number of patients	67	26 #
Proportion female sex	43.3%	44.0%	0.93
Mean age at treatment start date ##	63.2 (8.45)	65.9 (6.75)	0.16
Mean (mean rank) BMI at treatment start date ###	30.5 (29.3)	27.8 (27.5)	0.048 *
Patient Comorbidities
Active Smokers	11.9%	11.5%	1
Diabetes	19.4%	30.8%	0.24
Hypertension	47.8%	53.9%	0.60
Coronary Artery Disease	10.5%	3.9%	0.44
Patient ECOG Status
ECOG 0	60.5%	65.4%	0.67
ECOG 1	36.6%	34.6%
ECOG 2	2.99%	0
Tumor Stage
T1	0	3.85%	0.34
T2	7.46%	11.5%
T3	61.2%	61.5%
T4	31.3%	23.1%
Nodal Status
N1	62.7%	80.7%	0.094
N2	37.3%	19.3%

^# 16 prescribed four cycles and 10 prescribed 8 cycles; ^## Reported as ‘mean (SD)’; ^### Reported as ‘mean (median)’. * p ≤ 0.05.

). Overall, 16 CAPOX patients were prescribed four cycles of CAPOX, and 10 were prescribed eight cycles. We found that there was no significant difference between patient sex, age at treatment start date, or smoking status between FOLFOX- or CAPOX-treated patients. The average patient body mass index (BMI) at the start of treatment was higher in our population of FOLFOX-treated patients compared to CAPOX-treated patients (30.5 FOLFOX versus 27.8 CAPOX; p = 0.048). Comorbidities such as diabetes, hypertension, and coronary artery disease were not significantly different between treatment groups and similarly there was no difference found between patient performance status at the start of treatment. The T and N Stage of the resected tumors were not significantly different between groups. A breakdown of the specific tumor characteristics in each treatment group can be found in Supplementary Figure S1.

In both treatment groups over half of patients experienced diarrhea (

Table 2. Patient Toxicities Secondary to Adjuvant Treatment with FOLFOX or CAPOX.

	FOLFOX (%)	CAPOX (%)	p Value
Diarrhea (Grade 1 or 2)	50.8	65.4	0.20
Diarrhea (Grade 3)	2.99	26.9	0.002 **
Constipation (Grade 1 or 2)	32.8	15.4	0.092
Constipation (Grade 3)	0	0
Hand-Foot Syndrome (Grade 1 or 2)	7.46	19.2	0.11
Hand-Foot Syndrome (Grade 3)	2.99	0	0.52
Mucositis (Grade 1 or 2)	35.8	3.9	0.002 **
Mucositis (Grade 3)	1.49	0	0.72
Nausea/Vomiting (Grade 1 or 2)	38.8	61.5	0.048 *
Nausea/Vomiting (Grade 3)	4.48	11.5	0.34
Neuropathy (Grade 1 or 2)	83.6	69.2	0.12
Neuropathy (Grade 3)	2.99	0	0.52

* p ≤ 0.05. ** p ≤ 0.01.

). The frequency and grade of diarrheal symptoms varied by treatment regimen, with CAPOX-treated patients being more likely to experience grade 3 diarrhea compared to FOLFOX-treated patients (26.9% versus 2.99%; p = 0.002). Low grade nausea/vomiting episodes were more common in CAPOX-treated patients compared to FOLFOX (61.5% versus 38.8%; p = 0.048). Mucositis was more commonly associated with FOLFOX than CAPOX (35.8% versus 3.9%; p = 0.002) however nearly all the symptoms were of grade 1 or 2 severity. No statistically significant difference between treatments was observed with respect to the rates of constipation, hand-foot syndrome, grade 3 mucositis, grade 3 nausea, or neuropathy. Among the subgroups of individuals experiencing each toxicity, we recorded the median onset of that toxicity for each treatment group. In our population, the majority of toxicities presented within the first 1–2 treatment cycles (

Table 3. Treatment Cycle at the Onset of Chemotherapy-Induced Toxicity.

	FOLFOX	CAPOX (Whole Group)	CAPOX (4 Cycles)	CAPOX (8 Cycles)
Diarrhea	2 (2–4)	1 (1–1)	1 (1–1)	1 (1–1.8)
Constipation	1 (1–1.8)	1 (1–1.3)	1 (1–1.3)	None
Hand-Foot Syndrome	5.5 (3–10)	2 (2–2)	1.5 #	2 (2–2.5)
Mucositis	2 (1–4)	1 ##	1 ##	None
Nausea/Vomiting	1.5 (1–2)	1 (1–2)	1 (1–1)	2 (1–2)
Neuropathy	2 (2–5)	1 (1–1.8)	1 (1–2.3)	1 (1–1)

All values are reported as reported as median (25th percentile–75th percentile). ^# Mean/median of two cases. ^## Single case.

). This trend was consistent when CAPOX-treated patients were stratified by the number of prescribed cycles.

After assessing the proportion of patients in each treatment group that experienced each individual toxicity, the proportion of patients that experienced a particular grade of toxicity at any point during treatment was evaluated (

Table 4. Proportion of Patients Experiencing One or More Chemotherapy-Induced Toxicities by Severity.

	FOLFOX (%)	CAPOX (%)	p Value
Grade 1 toxicity	82.1	69.2	0.18
Grade 2 toxicity	71.6	57.7	0.20
Grade 3 toxicity	25.4	53.9	0.009 **
Grade 4 toxicity	1.49	0	0.720

** p ≤ 0.01.

). Grade 3 toxicities were significantly more common in CAPOX-treated patients compared to FOLFOX (53.9% versus 25.4%; p = 0.009). However, there was no significant difference between the proportion of patients that experienced grade 1, 2, or 4 toxicities between treatment regimens. Patient levels of hemoglobin, platelets, and neutrophils were also tracked through treatment. Neutropenia and thrombocytopenia were significantly more common in FOLFOX-treated patients compared to CAPOX, while rates of anemia did not appear to significantly differ between groups (

Table 5. Proportion of Patients Experiencing Treatment-Related Cytopenias.

	FOLFOX (%)	CAPOX (%)	p Value
Patients experiencing anemia during treatment (Males < 135 g/L, Females < 120 g/L)	62.7	57.7	0.66
Patients experiencing neutropenia during treatment (ANC < 1.5 × 103 neutrophils/mm3)	41.8	15.4	<0.001 ***
Patients experiencing thrombocytopenia during treatment (platelets < 100,000/µL)	43.3	11.5	0.004 **

** p ≤ 0.01. *** p ≤ 0.001.

).

We found that patients who received adjuvant treatment with CAPOX were significantly more likely to present to the emergency department secondary to treatment-related toxicity (mean 0.692 visits per patient versus 0.313 in FOLFOX patients, p < 0.001). We also observed that the mean number of hospitalizations secondary to CAPOX toxicity was greater in our population and was trending toward a statistically significant difference from FOLFOX-treated patients (

Table 6. Healthcare Resource Utilization Secondary to Chemotherapy-Induced Toxicity.

	FOLFOX	CAPOX	p Value
Number of visits to emergency department secondary to toxicity #	0.313 (0.0)	0.692 (1.00)	<0.001 ***
Number of hospital admissions secondary to toxicity #	0.209 (0.0)	0.385 (0.0)	0.066
Use of colony-stimulating factors during treatment	38.8%	3.85%	0.00100 **
Patients requiring transfusion during treatment	0	0
Total RBC units	0	0
Total platelet units	0	0

^# Reported as ‘mean (median)’. ** p ≤ 0.01. *** p ≤ 0.001.

). The use of colony-stimulating factors during adjuvant treatment was more common in the FOLFOX-treated group compared to CAPOX (38.8% of patients versus 3.85% of patients; p = 0.00100). In our population, none of the patients being treated with either regimen required transfusion of red blood cells (RBC) or platelets.

The starting dose intensity on cycle one of treatment was not significantly different between treatment regimens (

Table 7. Treatment Regime Adjustments Secondary to Chemotherapy-Induced Toxicity.

Regime Adjustments	FOLFOX	CAPOX	p Value
Dose intensity oxaliplatin on cycle 1 #	96.1 (100)	96.5 (100)	0.72
Dose intensity FU/capecitabine on cycle 1 #	96.1 (100)	96.2 (100)	0.99
Patients requiring treatment delay	58.2%	34.6%	0.041 *
Stoppage of treatment 2° to toxicity	10.5%	46.2%	<0.001 ***
Total oxaliplatin dose intensity received ##	73% (48.7)	66% (42.7)	0.341
Total 5-FU dose intensity received ##	88% (52.9)	69% (31.9)	<0.001 ***
Median cycle of first dose reduction ###	7 (4.75–9)	2 (2–4)
Median cycle at treatment stoppage ###	11 (6–11)	4 (2.25–5.50)

^# Reported as ‘mean (median)’; ^## Reported as ‘mean (mean rank)’; ^### Reported as median (25th percentile–75th percentile). * p ≤ 0.05. *** p ≤ 0.001.

). Once treatment had begun, we found that FOLFOX was associated with a greater proportion of patients requiring delays in treatment secondary to toxicity. However, despite a higher proportion of patients with treatment delays, many more FOLFOX patients compared to CAPOX patients were able to finish their prescribed treatment (46.2% versus 10.5%; p < 0.001) and FOLFOX patients completed a greater percentage of the total prescribed dose of 5-FU over their treatment course (88% versus 69%; p ≤ 0.001). There was no significant difference in the dose intensity of oxaliplatin that was received. Among patients that required dose reductions, the median cycle of the first dose reduction was cycle 7 (4.75–9) for FOLFOX and cycle 2 (2–4) for CAPOX-treated patients. Among patients that required complete stoppage of treatment, the median cycle for discontinuation among the FOLFOX group was the 11th (6–11) cycle, and the median among CAPOX-treated patients was the 4th (2.25–5.50) cycle.

3. Discussion

This study compared the two chemotherapy regimens that are in current clinical use for the adjuvant treatment of Stage III CRC in Newfoundland. In this paper we have provided a comprehensive evaluation of the frequency and severity of different toxicities as well as the real-world implications of these toxicities on treatment completion, patient safety, and healthcare resource utilization.

As part of our study, we performed a detailed audit of patient toxicities to determine not only the proportion of patients experiencing a given toxicity, but also the severity of symptoms. We found that while FOLFOX-treated patients had higher rates of grade 1 and 2 mucositis (Table 2), neutropenia, and thrombocytopenia (Table 5), CAPOX-treated patients had higher levels of grade 1 or 2 nausea and notably grade 3 diarrhea (Table 2). These findings are in agreement with the work of Loree et al. (2018) who also found that diarrhea and nausea were more common with CAPOX treatment while mucositis and neutropenia were more common with FOLFOX [11]. Loree et al. (2018) and Degirmencioglu et al. (2019) also reported an increased rate of hand-foot syndrome in CAPOX-treated patients, which is in line with the trend of our data [9,11]. Interestingly, a study performed by Mamo et al. (2016) found that nausea, diarrhea, and neuropathy were more common with FOLFOX than CAPOX chemotherapy [10]. However, their results recorded only the toxicity events that resulted in a dose reduction, whereas our work documented all toxicities and stratified them according to a verified scale [12].

An important finding of our study is that the overall proportion of patients experiencing a grade 3 level toxicity was far greater in CAPOX-treated patients (Table 4). In our population, greater than twice the proportion of CAPOX compared to FOLFOX patients experienced at least one grade 3 level toxicity. Grade 3 toxicities result in severe or medically significant pathology that limit self-care and result in hospitalization or a prolongation of admission [12]. This suggests that CAPOX patients are being placed at greater risk for serious adverse events compared to those taking FOLFOX, which reflects a reduced standard of care. In addition to safety concerns, the increased risk of these severe toxicities must be considered within the context of treatment compliance and continuation. A patient that is experiencing debilitating adverse events is unlikely to be adherent with oral capecitabine and far more likely to make the decision to forego the remainder of adjuvant treatment.

Citing the greater proportion of severe toxicities that were observed with the CAPOX regimen, we wanted to determine if these adverse events translated to increased utilization of healthcare services. We found that on average CAPOX-treated patients had visited the emergency room secondary to treatment-related toxicity with greater than twice the frequency compared to FOLFOX-treated patients (Table 6). Moreover, we observed a trend toward increased hospital admissions in CAPOX compared to FOLFOX-treated patients. To our knowledge, this is the first time that a clear increase in the utilization of emergency services has been demonstrated with CAPOX treatment. This vast difference means that CAPOX patients are likely being exposed to increased investigations, stress, and a reduced quality of life by factors that could be largely avoided with a switch to FOLFOX therapy. More emergency room visits and hospitalizations also place undue pressure on an already strained healthcare system and increases the exposure of immunocompromised patients to pathogens and unnecessary hospital-acquired infections.

We also evaluated the use of blood products during adjuvant treatment. As expected, given the increased rates of neutropenia in FOLFOX-treated patients (Table 5) we saw an increase in the use of colony-stimulating factors in the FOLFOX pool (Table 6).

As a final step in our evaluation, we wanted to independently determine if the differences in toxicities between FOLFOX and CAPOX have had any effect on the dosing schedule, likelihood of treatment completion, and the total dose intensity of treatment received by patients. We found that while treatment delays were more common in FOLFOX-treated patients, premature discontinuation of treatment was over four times more likely in patients prescribed CAPOX leading to an overall reduction in the 5-FU dose intensity received by CAPOX patients (Table 7). Additionally, due to the reliance on self-reports of adherence to oral capecitabine, the reduction in total 5-FU dose intensity received by CAPOX compared to FOLFOX-treated patients is likely even larger than reflected by our results.

Despite the changes in CAPOX-prescribing practices that were instituted by the IDEA trial and reflected in our data, our results were in line with previous literature that only evaluated patients prescribed 8 cycles of CAPOX. This was a critical finding because it confirms that patients are receiving a lower total percentage of their prescribed treatment even when the majority (16 of 26 in our study) of patients are being prescribed three months of CAPOX against a comparator of six months of FOLFOX. It is also likely that the increased rate of treatment delays in FOLFOX compared to CAPOX is inflated at least in part by the fact that many of the treatment intolerances experienced by CAPOX patients resulted in complete discontinuation of therapy rather than delay. This is important because clinical trials surrounding FOLFOX and CAPOX have based treatment efficacy on patients completing the prescribed regimen. It is unlikely that a patient who discontinues therapy prematurely will incur the same efficacy as a patient that is able to complete therapy. Given these results, CAPOX-treated patients may be at greater risk for disease recurrence.

One of the main strengths of this work was that it compared the treatment regimens according to current real-world prescribing practices. Since the 2018 release of the IDEA trial guidelines, many physicians have been prescribing 3-month CAPOX therapy for low-risk Stage III CRC. However, no previous studies had included these patients in the comparison with six months of FOLFOX. This study also provided a direct comprehensive comparison of individual toxicities and their severity using a standardized grading system [12]. Moreover, to ensure consistency in the grading of toxicities, all grades were assigned by a single individual based on physician and pharmacy notes.

A potential limitation of our study is that it was performed in a single province of Canada that is known to be influenced by a founder effect with high rates of hereditary cancer syndromes. Rates of dihydropyrimidine dehydrogenase deficiency (DPD) are also not known in the Newfoundland and Labrador population, as it is not routinely tested. This being so, the pharmacogenomics of this population may not be generalizable to all groups. This study also used a small sample size of only 93 participants. Larger studies will be necessary to validate our results. Finally, while the majority of demographic factors did not vary significantly between treatment groups, patients in the FOLFOX-treated cohort did have a higher average BMI compared to patients in the CAPOX group. While it is unlikely that these factors had any appreciable influence on our results, it is worth noting that differences in metabolic profile can affect pharmaceutical tolerance.

4. Materials and Methods

This project was conducted at the Dr. Bliss H. Murphy Cancer Centre in St. John’s which coordinates the care of all cancer patients across the province of Newfoundland and Labrador. The project was a multicenter retrospective cohort study and patients were sourced from four regions across the province (St. John’s, Corner Brook, Grand Falls-Windsor, and Gander). Site selection was based on the availability of patient data by electronic medical record. Patient records were obtained for all Stage III CRC patients from the outlined regions that were treated with adjuvant FOLFOX or CAPOX from 2015–2019. Due to feasibility of numbers, a simple random sample of 50% was taken from the pool of FOLFOX-treated patients prior to inclusion criteria screening. All CAPOX-treated patients were eligible for the study if they met the inclusion criteria. Patients from the resulting pools were included if they were over 18 years of age, had been pathologically staged as Stage III CRC, had no history of prior malignancy, and had undergone tumor resection with curative intent. Patients were also required to have started therapy within 12 weeks of their curative intent surgery. Patients were excluded from the study if they had received any radiation therapy, had switched therapy regimens for reasons other than treatment toxicity, or had received FOLFOX or CAPOX as part of a clinical trial.

At the time of this study, the FOLFOX regime being used by the Dr. Bliss H. Murphy Cancer Centre was mFOLFOX6 consisting of oxaliplatin 85 mg/m² provided I.V and leucovorin 400 mg/m² I.V concurrently over 2 h, followed by 5-FU 400 mg/m² I.V bolus and a continuous I.V infusion of 2400 mg/m² of 5-FU over 46 h, delivered every 14 days for 12 cycles. The CAPOX regime consisted of oxaliplatin 130 mg/m² I.V over 2 to 6 h on day 1 and capecitabine 1000 mg/m² orally every 12 h on day 1 to 14, delivered every 21 days.

Pathologic staging of patients was performed using the 8th edition of the American Joint Committee on Cancer Staging manual [13]. The performance status of each patient was assessed directly before the beginning of adjuvant therapy and was assigned according to the Eastern Cooperative Oncology Group performance status scale [14].

The category and grade of toxicities were assigned according to the Common Terminology Criteria for Adverse Events version 5.0 [12] and all toxicity data was sourced from physician notes and pharmacy cycle assessments that were performed during adjuvant treatment. If not already specified in the chart the grade of toxic events was assigned by a single independent source based on the description of the event provided in the chart data.

To directly compare the incidence of adverse events between treatment groups, patient demographics, chemotherapy-induced toxicities, and the severity of toxicities experienced by patients were reported as percentages of the total number of individuals in that treatment group (67 for FOLFOX, 26 for CAPOX). Healthcare resource utilization (hospitalizations or emergency department visits secondary to treatment-induced toxicity) was reported as mean events per person.

For the purposes of comparison, dose reductions were reported as the dose intensity of each pharmaceutical component. Total dose intensity was calculated as the total quantity of each chemotherapeutic agent received by the patient during the duration of treatment as a percentage of the total quantity that would be received if the patient completed the full standard chemotherapeutic regime. To account for prophylactic dose reductions made on the basis of age and comorbidities, the mean and median quantity of each chemotherapeutic agent given on cycle 1 was reported as a percentage of the standard cycle dosage for each regime.

All continuous and ordinal variables were assessed for normality using the Shapiro–Wilk test. Normally distributed data was compared using students t-test. The nonparametric Mann–Whitney U test was used for skewed data. For ease of reference, all continuous and ordinal variables were reported as means. Median or mean rank values as appropriate were also included in parentheses for all variables analyzed by the non-parametric Mann–Whitney U test. The X² test was used for categorical variables in which all expected values in the contingency table were >5. Fishers exact test was used when any cell in the contingency table had n < 5. All statistical analyses were performed using SPSS version 27.

5. Conclusions

In this study, we have demonstrated that FOLFOX and CAPOX have different toxicity profiles. We have shown that patients treated with CAPOX chemotherapy are more likely to experience toxicities of greater severity, require emergency services secondary to treatment-related toxicity, and to discontinue therapy compared to FOLFOX-treated patients. Treatment-induced toxicity has also resulted in CAPOX-treated patients receiving a lower total proportion of their prescribed 5-FU regimen. We hope that the evidence provided by this study helps physicians to make more informed treatment decisions for their adjuvant Stage III CRC patients.