Oral Capecitabine Exposures and Use of Uridine Triacetate: A 20-Year Retrospective Analysis

Abstract

Background and Objectives

Capecitabine is an oral prodrug of 5-fluorouracil. Toxicity can occur during therapy as well as acutely with overdose and particular genetic susceptibilities. Uridine triacetate is an effective antidote if given within 96 h of exposure. This study seeks to characterize accidental and intentional capecitabine exposures and uridine triacetate use, about which little has been published.

Methods

A retrospective review of capecitabine exposures from 30 April 2001 to 31 December 2021 reported to a statewide poison control center was performed. All single-substance oral exposures were included.

Results

In total, 81 of 128 reviewed cases were included, with a median age of 63 years. In total, 49 were acute-on-chronic exposures and 32 were acute exposures in capecitabine-naïve patients, 29 of which were accidental. Fifty-six (69%) were managed at home. Of these, none later recontacted the poison control center to report symptoms or were known to have later had healthcare facility evaluations. Of the 25 cases presenting for healthcare facility evaluation, 4 were acutely symptomatic. Thirteen were eligible for uridine triacetate, and six received it; no new or progressive toxicity was reported after. Three developed mild latent toxicity; otherwise, no morbidity or mortality was reported.

Conclusions

Accidental acute-on-chronic and acute ingestions of capecitabine appear to be well tolerated; most cases were managed at home. Unfortunately, little is known regarding the threshold at which toxicity may present following exposures. The threshold may vary individually given genetic susceptibilities. Management was heterogeneous, likely reflecting inadequate guidelines. Further research is needed to better delineate at-risk populations and treatment strategies.

Key Points

Accidental acute-on-chronic and acute ingestions of capecitabine appear to be well tolerated.

Management guidelines, especially regarding uridine triacetate use, are inadequate and have led to heterogeneous management.

Introduction

Capecitabine (Xeloda, Genentech, San Francisco, California, USA) is an oral prodrug of 5-fluorouracil (5-FU) introduced in 2001 that is currently used as a treatment for a variety of malignancies [1]. Capecitabine toxicity is well described and is most often reported during therapeutic treatment due to cumulative toxic effects. A potentially life-threatening acute toxicity can also occur following acute overdose and in patients with particular genetic susceptibilities [1–3].

In 2015, uridine triacetate (Vistoguard, Wellstat Therapeutics, Rockville, Maryland, USA) was introduced as the antidote for both capecitabine and 5-FU toxicity. Uridine triacetate has been shown to be highly effective if given as soon as possible, particularly within 96 h of exposure [3–6].

Despite years of clinical capecitabine and uridine triacetate use, little has been published regarding toxicity and management. This study seeks to characterize accidental and intentional exposures to capecitabine and the use of uridine triacetate.

Methods

A retrospective review was performed of capecitabine exposures reported to a statewide poison control center (PCC) from 30 April 2001 to 31 December 2021. All single-substance oral capecitabine exposures were included. Exclusion criteria were non-oral exposures, iatrogenic overdose, polydrug exposure, and unknown outcome. Patients on chronic capecitabine therapy calling in regard to dose timing or mild side effects were also excluded. The study was approved by our health system institutional review board.

Results

A total of 128 cases were identified for review. Following review, 81 were included (34 male patients and 47 female patients); 19 were excluded due to incorrect dose timing, 17 due to unknown outcome, 7 due to non-oral exposures, 2 due to therapeutic side effects, and 1 due to polydrug exposure. One case noted to be a duplicate entry was excluded.

The median age was 63 years (range 1–92 years), and 17 cases (21%) were under 16 years old. A total of 49 patients on active capecitabine therapy reported acute-on-chronic exposures. All were accidental extra doses with a median reported ingestion of 2000 mg (range 300–7500 mg). The median reported therapeutic dose was 1500 mg (range 150–3000 mg). A total of 32 capecitabine-naïve patients reported acute ingestions. Of these, 29 ingestions were accidental, with a median ingested dose of 500 mg (range 125–5000 mg), and 3 ingestions were intentional: a 14-year-old female patient who ingested 15,000 mg, a 19-year-old female patient who ingested an unknown quantity, and a 16-year-old male patient who ingested 7500 mg.

A total of 56 cases (69%) were managed at home. Of these, none later recontacted the PCC to report symptoms and none was known to have later presented for healthcare facility (HCF) evaluation.

In total, 25 cases presented for HCF evaluation. Four (16%) cases were symptomatic at the time of presentation: three with gastrointestinal upset and one with confusion, and nine (36%) cases were admitted, three to intensive care, with an average reported length of stay of 4.6 days (n = 8; range 1–10 days).

Of the 25 cases evaluated at HCFs, 13 presented after 2015 and thus were eligible for uridine triacetate therapy. In total, 6 (46%) of those 13 cases received uridine triacetate. Uridine triacetate was given in three cases due to acute gastrointestinal symptoms and in one case due to acute confusion. Two patients received uridine triacetate for asymptomatic post-exposure treatment. Two of the intentional ingestions, the 14-year-old female patient and the 19-year-old female patient, received uridine triacetate; the third intentional ingestion presented before uridine triacetate was available. Five of the six uridine triacetate treatments were completed; one was discontinued after 4 of the 20 total doses following an undetectable serum 5-FU concentration. For all six cases that received uridine triacetate, no new or progressive toxicity was subsequently reported. The case presenting with acute confusion rapidly improved after uridine triacetate administration.

Three cases, all initially asymptomatic, developed latent toxicity. A 2-year-old female patient developed vomiting approximately 4 h after accidentally ingesting 500 mg capecitabine, the aforementioned 16-year-old male patient developed palmar-plantar erythrodysesthesia 1 day after ingestion, and a 1-year-old male patient developed lymphopenia on outpatient follow-up labs drawn 4 days after accidental ingestion of 500 mg capecitabine. Otherwise, no morbidity or mortality was reported.

Discussion

Capecitabine is activated to 5-FU in vivo by a three-step enzymatic process; the final step is the formation of 5-FU by thymidine phosphorylase [1, 2]. While capecitabine and 5-FU produce the same toxicity, capecitabine is different in two important ways. First, the oral formulation allows for home administration, and therefore the increased possibility of dosing errors and nontherapeutic exposures. Second, the activation of capecitabine to 5-FU is intended to be more tumor selective compared with direct 5-FU administration, as target tumors have higher thymidine phosphorylase levels than normal cells [1, 2]. The 5-FU is then further metabolized into several cytotoxic metabolites. These metabolites are directly incorporated into DNA and RNA as ineffectual nucleotides. One metabolite, fluorodeoxyuridylate (FdUMP), also inhibits thymidylate synthase, which is necessary for thymidine and therefore DNA synthesis. The cumulative cytotoxic effect triggers apoptosis of the target cell [2]. Figure 1 details capecitabine metabolism.

Fig. 1.

Capecitabine metabolic pathways [1, 2, 16–18]

Approximately 20–30% of capecitabine patients experience some form of drug toxicity during routine treatment; renal dysfunction can increase the risk and severity. This toxicity usually occurs within weeks of initiating treatment and can range in severity from mild to life-threatening. Toxicity can also be cumulative across treatment cycles. Common clinical presentations include nausea, vomiting, diarrhea, stomatitis, and palmar–plantar erythrodysesthesia. Anemia, cytopenias, and hyperbilirubinemia are commonly encountered laboratory abnormalities [1, 7, 8]. Toxicity is usually managed with dose adjustment or discontinuation by the patient’s oncologist. Uridine triacetate use is not routinely recommended in this circumstance as it may decrease the efficacy of capecitabine [4, 6, 8].

However, capecitabine can also cause an acute toxicity, which usually occurs within days of exposure and can be life-threatening. Gastrointestinal symptoms, stomatitis, cytopenias, and dermatotoxicity are generally more severe than with routine therapeutic use toxicity. Prominent neurotoxic and cardiotoxic effects can also develop. This acute toxicity occurs in two distinct circumstances. The first follows therapeutic initiation of capecitabine and is known as “early-onset” toxicity. Early-onset toxicity is due to patient-specific factors such as enzyme functional abnormalities or deficiencies that reduce 5-FU catabolism and precipitate toxicity at therapeutic dosing. Acute overdose produces the same clinical findings as early-onset toxicity by a different mechanism. At this time, the toxic overdose threshold remains poorly characterized [4, 7, 9, 10].

Dihydropyrimidine dehydrogenase (DPD) is the primary enzyme responsible for 5-FU catabolism. DPD deficiency is a known precipitant of early-onset toxicity. Similarly, polymorphisms in thymidylate synthase and cytidine deaminase can increase the risk of toxicity. An acute overdose, meanwhile, saturates normally functioning DPD, resulting in 5-FU accumulation. [1–3, 7, 11–13]. Figure 2 details capecitabine catabolism.

Fig. 2.

Capecitabine catabolic pathways [1, 2, 16–18]

Uridine triacetate is the treatment of choice for early-onset toxicity from both capecitabine and 5-FU. It is also recommended in the setting of acute overdose, even if asymptomatic at the time of presentation [4, 6, 7, 9, 10]. Pooled analysis of two clinical trials of uridine triacetate for the treatment of capecitabine and 5-FU overdose showed 97% (114/117) of these cases survived, while a smaller cohort of cases with early-onset toxicity showed 89% (16/18) survived if uridine triacetate was administered within 96 h of the last capecitabine or 5-FU dose, leading to a 96% (130/135) overall survival rate. A total of 4 of the 135 patients were treated beyond the 96-h window, and 2 died (50% survival). These results were compared with 25 historical 5-FU overdose cases treated only with supportive care, which had a 16% (4/25) survival rate [14]. Later case reports have supported uridine triacetate as an effective treatment [9, 10].

Unfortunately, little is known about the threshold at which toxicity may present following acute or acute-on-chronic capecitabine exposures. The threshold may vary from person to person given the aforementioned susceptibilities. Precise guidelines on when uridine triacetate is indicated are lacking, particularly in the setting of capecitabine overdose regardless of symptoms.

Our analysis revealed several important insights. Accidental acute-on-chronic and acute ingestions of capecitabine were the most common types of exposures and appear to be well tolerated. Intentional acute ingestions were rare. Most cases were managed at home. Per PCC protocol, all patients managed at home were advised to call back if any symptoms occurred, and none did. Most of those who presented to HCF were asymptomatic. Only one patient developed more than minor symptoms, presenting with acute confusion. A majority of patients were discharged from the emergency department, and no return visits were reported to our Poison Center. Further, most acute-on-chronic ingestions were accidental double doses that had no reported ill effects. Finally, management of capecitabine exposures, in particular utilization of uridine triacetate, is highly variable. In total, 6 of 13 eligible cases received uridine triacetate. It is unclear why the other seven cases that presented after 2015 did not. Only a minority of patients were admitted, some to intensive care, but less than half of those admitted were symptomatic at the time of admission.

Our study is limited by several factors, primarily by the retrospective design and use of PCC data. Our PCC relies on communication by phone from providers, hospitals, and the general public at the time of the call(s). Outside electronic medical records are not available to correct, verify, or augment the information contained in the PCC database. Consequently, recorded information is vulnerable to erroneous reporting and recollection, but cannot be later verified or corrected. Missing data points noted during retrospective review also cannot be readily obtained [15]. It is further limited by completeness of follow-up, especially with cases managed at home. While all of these cases were advised by protocol to contact the Poison Center if any changes occurred, none was directly followed up by PCC staff. We assume in our analysis that lack of additional contact with our PCC, either directly or from HCFs, meant the patient did not develop subsequent toxicity. However, given that we did not routinely initiate follow-ups on most cases, we cannot say with certainty that this was the case. Finally, our sample size, especially of patients treated with uridine triacetate, was small and limits the power and generalizability of our analysis. In particular, we cannot draw specific conclusions regarding intentional ingestions due to rarity.

Conclusions

Incidental capecitabine exposures, such as accidental double doses or small single dose ingestions, appear to be generally well tolerated. Only a small minority were known to have developed symptoms, most of which were minor. A majority were managed without HCF referral. Management of those who presented for HCF evaluation was heterogeneous, especially in terms of uridine triacetate use. We feel this reflects the lack of detailed clinical guidance available for patients presenting in these circumstances. Further research is needed to better delineate at-risk populations and empiric treatment strategies.

Declarations

Funding

Not applicable.

Conflict of interest

No relevant conflicts of interest.

Ethics approval

All procedures in this study were in accordance with the 1964 Declaration of Helsinki (and its amendments), and the details of the institutional review board that approved the study.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Availability of data and material

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Code avaialbility

Not applicable.

Author contributions

JS was responsible for study conception, data collection and analysis, and manuscript preparation and review; NF, JH, and HG were responsible for manuscript preparation and review; FLC was responsible for study conception, data collection and analysis, and manuscript preparation and review; and AM was responsible for study conception, manuscript preparation and review, and overall study supervision as primary investigator.