Yap, Kevin Yi-Lwern, Chan, Alexandre, Low, Xiu Hui and Chui, Wai Keung. (2010) Clinical predictors of chemotherapy-induced nausea and vomiting through identification of principal variables in patients undergoing AC and XELOX regimens. Supportive Care In Cancer, Vo.18 (Suppl.3). S142-S142. ISSN 0941-4355

Abstract

Objectives: This study aims to identify the clinical predictors of Chemotherapy-Induced Nausea and Vomiting (CINV) in patients undergoing moderately-emetogenic chemotherapy (MEC) through a novel strategy of identifying Principal Variables (PVs) using Principal Components (PCs) analysis. Methods: NV events were collated from 332 cancer patients undergoing AC (doxorubicin 60mg/m2 and cyclophosphamide 600mg/m2 on day 1, every 14 or 21 days) or XELOX (oxaliplatin 130mg/m2 on day 1, every 21 days, and capecitabine 2000mg/m2/day on days 1–14). These events were correlated with 16 variables representing various CINV predictors (anxiety, chemotherapy-related, drinking habits, fatigue, gender, labyrinthitis, concurrent radiotherapy, and histories of CINV, motion and morning sicknesses). PC analysis, a mathematical projection technique for explaining causes of variation in datasets, was used for analysis. PVs were extracted based on their highest weights on the PCs. Results: Majority of patients were Chinese and over 50 years old. Higher proportions of patients in both regimens suffered from delayed than acute NV. Common PVs that were observed in patients with both acute and delayed NV included ‘ex-drinkers’ (17.8–21.4% variation), ‘history of morning sickness’ (7.3–11.2% variation) and ‘concurrent radiotherapy’ (6.2–6.9% variation). On the other hand, ‘gender’ explained 14.1–16.0% of the variation in patients suffering from both acute and delayed vomiting, while ‘non-drinkers’ explained 13.4 13.7% of the variation in patients with acute and delayed nausea. In contrast, ‘histories of nausea and vomiting’ only contributed to the variation (5.5–9.0%) in the acute phase of NV. Similarly, ‘earache or ringing’ contributed to 7.1–10.4% of the variation in delayed NV. Conclusions: The opportunity for PC analysis to identify various clinical predictors in CINV patients on MEC regimens is indeed attractive. However, several key issues still need to be addressed before it finds widespread applications in CINV prediction for a variety of cancer treatment regimes.

Request changes to a record

Actions (login required)

View Item