Abstract
BACKGROUND: The clinical impact of hypomagnesemia induced by necitumumab plus gemcitabine and cisplatin (GCN) as a second-line or later therapy is unclear. OBJECTIVE: We aimed to evaluate the clinical characteristics and survival impact of hypomagnesemia induced by this therapy. DESIGN: This was a sub-analysis of the retrospective multicenter NINJA study. METHODS: Among the 93 patients enrolled in the NINJA study, this subanalysis included 75 patients with baseline serum magnesium concentrations. RESULTS: The incidence of grade ⩾2 hypomagnesemia was 18.0% in the patients with normal baseline serum magnesium concentrations and 42.8% in those with low concentrations (p = 0.073). The discontinuation rates of GCN treatment owing to hypomagnesemia in each group were 0% and 7.1%, respectively (p = 0.187). The number of necitumumab doses and severity of hypomagnesemia were positively correlated (r = 0.389, p < 0.001). Patients who developed hypomagnesemia in fewer than 21 days after the first dose of GCN (n = 12) had significantly poorer progression-free survival (PFS) than those without the condition (n = 63; median: 4.1 vs 4.4 months, p = 0.048). A similar trend was observed for OS (median: 9.7 vs 15.7 months, p = 0.062). These results were maintained after multivariate analyses (PFS: hazard ratio (HR) 2.46, p = 0.014; OS: HR 2.78, p = 0.021). CONCLUSION: GCN as a second-line or later therapy may be tolerable regardless of the patient's baseline serum magnesium concentration. On the other hand, early serum magnesium reduction with this therapy is associated with a poor prognosis. However, caution should be needed because our results lacked sufficient information for confounding variables other than those analyzed here that may influence the correlation between hypomagnesemia and survival.