Investigating the Molecular Epidemiology and Prognosis of Smoking and Gliomas
Gabby Bognet1, Sarah Ahr2, Shakti Ramkissoon3, Michael Chan5, Christina Cramer5, Stephen Tatter6, Adrian Laxton5, Jaclyn White5, Glenn Lesser4, Roy Strowd7
1Wake Forest University, 2Thomas Jefferson University, 3Pathology, 4Neuro-Oncology, Wake Forest Baptist Health, 5Wake Forest Baptist Health, 6Wake Forest School of Medicine, 7Wake Forest School Of Medicine
Objective:

To evaluate the association between smoking status, genetic alterations in primary glial neoplasms, and impact on survival. 

Background:
Cigarette smoking may contribute to cancer-associated genomic instability with increased tumor mutation burden and frequency of KRAS, TP53, and other mutations. Few studies have explored smoking-associated genomic instability in patients with primary gliomas.
Design/Methods:

A retrospective analysis was performed on adult patients (>18 years) diagnosed with a malignant glioma (i.e., astrocytoma, oligodendroglioma, glioblastoma; WHO Grade 2-4) between 2000-2020 and receiving primary oncologic care at Wake Forest Baptist Comprehensive Cancer Center. Data was queried for demographic, clinical, molecular, and treatment characteristics. Smoking status was defined as active/former vs never smoker.

Results:

291 patients were identified; mean age 59+15 years; 54% male; 158 GBM, 64 astrocytoma (28 LGA, 36 AA), 48 oligodendroglioma (39 LGO, 10 AO), 20 other glioma. Of these, 58% were never smokers and 42% were current/former smokers with 7.1+12 years mean smoking duration. There was no difference in age (p=0.38), diagnosis (0.07), extent of resection (0.75), prior radiation (0.08), or chemotherapy (0.75) by smoking status. The most frequently altered genes were pTERT (69%), CDKN2A (41%), TP53 (45%), CDKN2B (38%), EGFR (33%), IDH (31.0%); mutational profiles did not differ by smoking status for all gliomas. In patients with astrocytoma, 73% of smokers and 42% non-smokers had TP53 mutation (p = 0.10); 7% smokers and 42% non-smokers had EGFR amplification (p = 0.06). Median OS was shorter for smokers (46 vs 141 months, p=0.045) with 42% greater risk of death (HR 1.42, 95%CI 1.01-1.99, p=0.045). This remained when controlling for histologic diagnosis (adjusted HR 1.45, 1.02-2.04, p=0.034) but not for age and diagnosis (adjusted HR 1.25, 0.88-1.78, p=0.20).

Conclusions:

Smokers have shorter survival from malignant glioma. While differences in age may contribute, smoking-related genomic alterations are not a mechanism for survival differences.

10.1212/WNL.0000000000202193