Retrospective analysis of the impact of dose delay and reduction on outcomes of colorectal cancer patients treated with FOLFIRI‑based treatment

Background

Colorectal cancer (CRC) is one of the leading causes of cancer death worldwide. Survival among patients with metastatic colorectal cancer (mCRC) has been improved since the addition of monoclonal antibodies to vascular endothelial growth factor (VEGF) or endothelial growth factor receptor (EGFR). Irinotecan and oxaliplatin added to 5-fluorouracil/leucovorin therapy remain the key drugs in the treatment of patients with mCRC (Innocenti et al., 2021). At present, irinotecan‑based chemotherapies combined with cetuximab or bevacizumab are standard treatment regimens for mCRC (Elez, Argilés & Tabernero, 2015).

The importance of a high relative dose intensity (RDI) for survival has been established for some cancers, including breast cancer (Lyman, Dale & Crawford, 2003), ovarian cancer (Anuradha et al., 2016) and lymphoma (Lyman et al., 2004); however, information about the effects of a high RDI among patients with CRC is limited. A high RDI is associated with improved survival, especially for patients with breast cancer or non-Hodgkin’s lymphoma (NHL) (Lyman, Dale & Crawford, 2003; Lyman et al., 2004). An RDI below 70–85% is generally considered to be a clinically meaningful threshold for reduction in the RDI based on previously published studies (Hanna et al., 2013; Lyman, 2009; Aspinall et al., 2015; Crawford et al., 2020). In the adjuvant setting, the importance of receiving full-dose chemotherapy on time has been well established (Aspinall et al., 2015; Gray et al., 2019; Denduluri et al., 2015). However, clinical goals differ from adjuvant settings, which include control of cancer growth, symptom management and extension of life in advanced disease. Chemotherapy dose delay and reduction are common in advanced disease to manage haematologic and infectious side effects or surgery (Crawford et al., 2020; Denduluri et al., 2018). In addition, there are nonmedical indications for delays, such as holidays, poor compliance and personal schedule preferences (Kogan, Davis & Brooks, 2019). Such agents are commonly associated with toxicities, leading to the occurrence of dose delay and reduction, which contribute to reduced RDI.

Dose delay and reduction occur more frequently in real-world populations than in more selected clinical trial populations, in which the benefits of these regimens were initially tested. However, the impact of chemotherapy modification on survival has not been well confirmed in the real-world setting. In addition, there are no established recommendations for delay time or dose reduction. Therefore, the purpose of this analysis was to retrospectively analyse the effect of dose delay/reduction on progression-free survival (PFS) and overall survival (OS) among patients with mCRC receiving FOLFIRI as first-line chemotherapy.

Methods

Results

Effect of chemotherapy dose delay on chemotherapy efficacy

There were 114 patients with chemotherapy dose delay (dose delay >3 days) and 30 patients without chemotherapy dose delay (dose delay ≤3 days). Between the two groups, there was no statistically significant difference in OS (p = 0.37). However, the PFS of patients without a dose delay was better than that of patients with a dose delay (p = 0.002) (Fig. 1).

Effect of 5-Fu dose reduction on chemotherapy efficacy

We used dose reduction and RDI to evaluate the effect of 5-Fu dose reduction on prognosis. As shown in Figs. 2A and 2B, PFS and OS were both better in patients without 5-Fu dose reduction than in patients with 5-Fu dose reduction (p < 0.001 and p = 0.024, respectively). However, there were no differences in OS for both 5-Fu dose reduction and RDI. As shown in Figs. 2C and 2D, only PFS was better in patients with a high 5-Fu RDI than in patients with a low 5-Fu RDI (p < 0.001). However, there was no difference in OS between the high 5-Fu RDI group and the low 5-Fu RDI group (p = 0.23).

Stratified analyses of OS and PFS according to 5-Fu RDI

In the stratified analysis of OS based on the 5-Fu RDI, we found that in the age <65 years subgroup, patients with a high RDI had better OS. However, in the age ≥65 years subgroup, patients with a low RDI had better OS (Fig. 3A). In the stratified analysis of PFS based on the 5-Fu RDI, patients with a high 5-Fu RDI had better survival in almost all subgroups. However, in the age ≥65 years subgroup, patients with a low RDI tended to have a better PFS (Fig. 3B).

Effect of 5-Fu dose reduction on chemotherapy efficacy in young and old patients

Patients younger than 65 years were defined as the young cohort, and those older than 65 years were defined as the older cohort. In the young cohort, PFS and OS were both better in patients with a high 5-Fu RDI than in those with a low 5-Fu RDI (p = 0.005 and p < 0.001, respectively) (Figs. 4A and 4B). In the older cohort, patients with a low 5-Fu RDI had better OS than those with a high 5-Fu RDI (p = 0.025, Fig. 4C). There was no statistically significant difference in PFS for 5-Fu RDI (Fig. 4D).

Effect of irinotecan dose reduction on chemotherapy efficacy

We also used dose reduction and RDI to evaluate the effect of irinotecan dose reduction on prognosis. As shown in Fig. 5, there were no statistically significant differences in either PFS or OS for either dose reduction or RDI.

Discussion

The optimal chemotherapy regimen and treatment of mCRC have been the subject of many retrospective and prospective studies in the past decade. Due to personal or disease reasons and treatment-related effects, it is common to modify the chemotherapy dose and interval of the chemotherapy cycle. Incidences of dose delays, dose reduction, missing doses, and reduced RDI were higher in those with CRC than in those with breast cancer, NHL and non-small cell lung cancer (NSCLC) (Denduluri et al., 2015). Reduced RDI was common among old patients with CRC who received adjuvant or neoadjuvant chemotherapy. Older patients may have poor tolerance and compliance to chemotherapy (Denduluri et al., 2015). However, few studies have focused on the impact of chemotherapy dose modification in metastatic CRC patients. To our knowledge, this is the first study to analyse the correlation between first-line chemotherapy dose or schedule modification and survival in mCRC patients.

In clinical practice, dose reduction and delay of cytotoxic agents usually lead to a reduction in RDI, resulting in adverse events. In our research, a delay of more than 3 days was associated with significantly poorer PFS, leading to a significantly increased risk of disease progression. However, there was no statistically significant difference in OS. Haematological toxicity was the most common reason for dose delays (Lyman, Abella & Pettengell, 2014). Our findings are consistent with the findings of similar studies conducted in other cancer sites (Joseph et al., 2015; Starbuck et al., 2018). In many previous studies, dose delay was defined as chemotherapy administration delay ≥3 or ≥7 days from the intended date (Crawford et al., 2020; Denduluri et al., 2015). It is likely that if more prolonged cut-offs were utilized, the adverse effect of dose delay would have been greater. Our current data demonstrated the importance of avoiding dose delay if at all possible.

In this study, we evaluated the relationships between the RDI of irinotecan and 5-Fu and survival outcomes, specifically the impact of dose reduction on PFS and OS in patients receiving FOLFIRI. We found that the dose reduction and RDI of 5-Fu might have a stronger association with survival than that of irinotecan. Patients with mCRC who received an RDI > 80% of irinotecan had no significant difference from those who received an RDI ≤ 80% in either PFS or OS. However, we found that a 5-Fu dose reduction >85% was a significant predictor of PFS and OS. Furthermore, a higher RDI of 5-Fu had a significant positive impact on PFS. Therefore, we conclude that irinotecan dose reduction may have less impact on survival than 5-Fu for patients receiving FOLFIRI-based first-line chemotherapy. If drug dose modification is carried out, the dose of irinotecan may be reduced first, rather than 5-Fu. However, it is important to be cautious in dose reduction because the number of patients enrolled in our study was too small. Additionally, doctors and patients should try to avoid extending the chemotherapy interval. When the chemotherapy is delayed, it will become a 3-week regimen. This issue should be further explored in a larger dataset.

In contrast to our results, a study from Japan found that a higher RDI of irinotecan had a significant positive impact on the objective response rate (ORR), disease control rate (DCR), PFS and OS. Dose reduction of irinotecan was an independent prognostic factor for PFS, while dose delay was not significantly associated with PFS (Nakayama et al., 2014). However, we found that reducing the RDI of irinotecan did not affect the curative effect on tumour response and survival outcomes. Previous evidence suggests that polymorphisms homozygous for UGT1A1 might be associated with increased incidences of serious haematological or gastrointestinal toxicities (Satoh et al., 2011). Several dose escalation studies found that the dose of irinotecan should be guided by the UGT1A1 genotype for patients with gene mutations to reduce the dose without affecting the effect of irinotecan (Xu et al., 2018). Dose modification of irinotecan based on UGT1A1 polymorphisms might explain these results. However, most studies showing adverse consequences of dose modification were basically retrospective, and the sample size was limited.

Some studies have confirmed that age-related comorbidities (Delgado-Ramos et al., 2020), physiological changes, female sex, obesity with a body surface area (BSA) >2.0 m² and the progressive loss of systemic protein or muscle (Cespedes Feliciano et al., 2017) will increase the risk of chemotherapy-induced toxicity. Many studies have found that older age was a risk factor for the development of chemotherapy-associated toxicities (Neugut et al., 2006; Søgaard et al., 2013; Shayne et al., 2007, 2009). Chemotherapy-induced neutropenia usually leads to hospitalization and the need for intravenous antibiotics. Neutropenia may result in dose reduction, delay, or even discontinuation of chemotherapy, which in turn may affect the prognosis of patients (Lyman, Abella & Pettengell, 2014). Our research revealed that 5-FU dose adjustment has a significant impact on prognosis and survival. Moreover, PFS was significantly different between low and high RDI of 5-FU. However, there were no differences in OS. From subgroup analysis, patients in the low 5-Fu RDI group were associated with poor PFS and OS in the young age (<65 years) population. However, this difference was not observed in the older group (≥65 years). Increased age and comorbidities, as well as the toxicity of 5-FU, may also lead to a reduction in chemotherapy completion, and a shorter duration of chemotherapy is related to poorer survival in mCRC. This tells us that chemotherapy dose delay and reduction need to be performed with caution in young patients.

Studies that evaluated the impact of dose reduction and delays on patient outcomes have been inconsistent to date. Research by Starbuck et al. (2018) and Olawaiye et al. (2018) showed that on-time completion of chemotherapy is associated with improved survival and higher complete response rates in epithelial ovarian cancer (EOC). A delayed completion of chemotherapy was associated with decreased survival. This view is similar to our research. However, the study did not consider the effect of dose modification on prognosis. However, some studies investigated the effect of dose reduction and/or delay and found no significant difference in survival for any group in EOC (Nagel et al., 2012; Sivakumaran et al., 2020). Due to the limited sample size of this study, we cannot provide more convincing information. In addition, a retrospective study explored whether chemotherapy delay alone was associated with decreased survival in the HER2-positive breast cancer subtype (Liutkauskiene et al., 2018). This finding establishes the importance of molecular typing and suggests that it may have an impact on the results. We expect larger sample data to explore the correlation between chemotherapy delay and RAS gene status. Additionally, we may have overlooked the impact of targeted drugs on prognosis and survival. Although the patients who have been successfully treated by resection of metastases or who have benefited from local treatment and radiotherapy have experienced chemotherapy delay, the above treatments have improved the prognosis of the patients. The benefits of these treatments may interfere with the correlation between RDI and prognosis. Further work and large samples are needed to elucidate the correlation between RDI and survival.

In conclusion, our study confirms the importance of maintaining the schedule of chemotherapy regimens. In the real world, it is important to reduce the occurrence of chemotherapy delay as much as possible. We found that the RDI of 5-Fu might have a stronger association with survival than that of irinotecan. Especially among patients younger than 65 years, 5-Fu doses need to be modified with caution. Further work is needed to clarify the role of dose delay and reduction in chemotherapy administration.

Supplemental Information