Palmar-plantar erythrodysesthesia in patients receiving capecitabine and intratumor thymidine phosphorylase and dihydropyrimidine dehydrogenase: is there a pharmacologic explanation?

BACKGROUND

Palmar-plantar erythordysesthesia (PPE) is the most common toxicity of capecitabine. Preclinical studies have shown that radiation therapy (RT) upregulates thymidine phosphorylase (TP), which could in turn increase the efficacy of capecitabine. Capecitabine is degraded by dihydropyrimidine dehydrogenase (DPD), and a deficiency in this enzyme can increase the toxicity of capecitabine. However, the effect of radiation therapy on frequency of PPE and association with TP and DPD has not been fully characterized.

PATIENTS AND METHODS

Toxicity data were collected prospectively in 33 patients and retrospectively in 25 patients with locally advanced pancreatic cancer enrolled in 3 clinical trials who received capecitabine/RT followed by capecitabine alone. Tumor specimens on 33 patients were procured via endoscopic ultrasonography standards 1 week befoer RT and 2 weeks after RT to evaluate TP and DPD messenger RNA levels. Roche grading was used to assess PPE.

RESULTS

Among 58 patients, 14 (24%) developed PPE. The capecitabine group tended to have a higher incidence of PPE than the capecitabine/RT group (17.2% vs. 10.3%; P = .12). Grade 2/3 PPE, was observed in 15.5% of patients receiving capecitabine and 1.7% of patients receiving capecitabine/RT (P = .0078; overall: 17.2%). Median cumulative dose of capecitabine for first development of PPE was 235,000 mg/m2 in the capecitabine/RT group, and 3,185,000 mg/m2 in the capecitabine group, with the relative frequency of an event occurring in the capecitabine/RT arm versus capecitabine of 0.59. Time to occurrence of first episode of PPE was a median of 5 weeks with capecitabine/RT and 6 weeks with capecitabine. Log-rank test showed that neither age, sex, performance status, nor ethnicity were associated with development of PPE. There was no clear difference in tumor responses of patients who had PPE versus those who did not have PPE. Mean tumor TP level was higher among patients with versus without PPE (275.77 vs. 215.29; P = .32). Mean tumor DPD level was lower among patients with versus without PPE (55.18 vs. 63.58; P = .49). Mean TP-DPD ratio was higher among patients with versus without PPE (10.29 vs. 3.04; P = .31).

CONCLUSION

No significant association of PPE with higher tumor thymidine phosphorylase or lower tumor DPD levels was found. There was no difference in the incidence, severity, or time to occurrence of PPE with capecitabine/RT versus capecitabine, indicating no effect of RT. Further studies to evaluate the underlying mechanism of PPE are required.