Association Between Pre-existing Schizophrenia and Lung Cancer Outcomes: A TriNetX Analysis.

INTRODUCTION

Patients with schizophrenia and related disorders have shorter life expectancies than their peers. While cancer incidence rates are similar between patients with and without schizophrenia, few studies have investigated lung cancer mortality rates in patients with baseline schizophrenia-spectrum disorders (SSD). This study compares all-cause lung cancer mortality rates between patients with and without SSD.

MATERIALS AND

METHODS

We queried the TriNetX Network for patients with primary lung cancer. Two cohorts were constructed, reflecting a diagnosis of schizophrenia and related disorders preceding a lung cancer diagnosis. The following International Classification of Diseases codes were included: schizophrenia (F20), schizotypal disorder (F21), delusional disorder (F22), schizoaffective disorder (F25), and primary lung cancer (C34). Patients were 1:1 propensity-matched. Kaplan-Meier survival curves were generated at 1, three, and 5 ys.

RESULTS

Before propensity matching, 86,667 (no-SSD) and 514 (SSD) patients were identified. After propensity matching, each group contained 511 individuals. Differences in all-cause lung cancer mortality were found in SSD patients at 1 (hazard ratio (HR) 0.790, P = 0.0364), three (HR 0.760, P = 0.0034), and five (HR 0.789, P = 0.0082) y. When analyzed by stage, only stage 3 had statistically worse survival at three and 5 ys. Median survival in the SSD and no-SSD groups was 641 and 1132 ds, respectively.

CONCLUSIONS

Lung cancer outcomes in SSD were found to be worse. In this matched cohort, differences in outcomes were observed alongside variations in treatment options chosen, with the largest differences observed in stage 3. However, further research accounting for treatment modalities, treatment adherence, and cause of death is warranted.