Comparison of oxaliplatin-related hepatic sinusoidal injuries in patients with gastric cancer who received S-1 or capecitabine as a combination chemotherapy

Abstract

Abstract Comparison of oxaliplatin-related hepatic sinusoidal injuries in patients with gastric cancer who received S-1 or capecitabine as a combination chemotherapy

Background: Oxaliplatin-based chemotherapy causes hepatic sinusoidal obstruction syndrome (SOS), resulting in portal hypertension with splenomegaly (SM). We herein aimed to compare the severity of hepatic SOS, and its impact on chemotherapy dose reduction or delay between two of the most commonly used oral fluoropyrimidines, S-1 and capecitabine, when combined with oxaliplatin in patients with gastric cancer.

Methods: We analyzed patients from two prospective trials for gastric cancer, adjuvant XELOX (capecitabine 1000 mg/m2 bid on D1–14 + oxaliplatin 130 mg/m2 on D1 q3w, eight cycles; n = 52) and palliative SOX (S-1 40 mg/m2 bid on D1–14 + oxaliplatin 130 mg/m2 on D1 q3w, continuous [SOX-C, n = 52] vs. intermittent [SOX-I, discontinuing after 6th and restarting on progression, n = 53]). Spleen volume was measured using the Rapidia software.

Results: Baseline sex, age, Eastern Cooperative Oncology Group Performance Status, platelet/liver enzyme/bilirubin levels, and spleen volume did not significantly differ between the XELOX and SOX-C groups. After the eight cycles, the SOX-C group showed more SM and hepatic parenchymal heterogeneity on CT scan and more severe hyperbilirubinemia and thrombocytopenia than the XELOX group. SM was significantly more associated with S-1 combination therapy (63.5% after eight cycles of SOX vs. 26.9% after the eight cycles XELOX; p < 0.001). The incidence of thrombocytopenia (≤ 100 × 103/μL) was significantly higher in SOX-C than in XELOX after six cycles (n = 20, 38.5% vs. n = 8, 15.4%; p = 0.008). In the SOX-C group, 209 of 1288 cycle treatments (16.2%) were delayed or reduced among 23 of 52 patients (44.2%) due to the thrombocytopenia and liver function test abnormalities, whereas in the XELOX group, only 28 of 832 cycles (3.3%) among four of 52 patients (7.6%) were delayed or reduced with same causes (p = 0.001).

Conclusion: S-1 apparently worsens oxaliplatin-induced hepatic sinusoidal injuries more than capecitabine in gastric cancer patients. This could increase the incidences of splenomegaly, thrombocytopenia, bilirubin increase, and dose reduction or delay of chemotherapy.