Lower Tumor Mutational Burden is Associated with Faster CNS Progression in Patients with Resected Brain Metastases Who Undergo Checkpoint Inhibitor Therapy

Sunday, May 5, 2024

1:47 PM – 1:49 PM CT

Location: S405a

Presenting Author(s)

Ramin Morshed, MD

Fellow
Mayo Clinic
San Francisco, California, United States

Co-Author(s)

Poojan Shukla, n/a

Medical Student
UCSF
San Francisco, California, United States
Manish Aghi, MD, PhD

PI
UCSF
San Francisco, California, United States
Mitchel S. Berger, MD (he/him/his)

UCSF BTRC Director, Professor of Neurosurgery
UCSF
San Francisco, California, United States
Shawn Hervey-Jumper, MD

Associate Professor
Department of Neurological Surgery, University of California, San Francisco
San Francisco, California, United States
MN

Minh Nguyen, BS

Medical Student
UCSF, United States
HV

Harish Vasudevan, MD PhD

Assistant Professor
UCSF, United States
Jacob S. Young, MD

Resident
UCSF
Charlottesville, Virginia, United States

Disclosure(s):

Ramin Morshed, MD: No financial relationships to disclose

Mitchel S. Berger, MD: No financial relationships to disclose

Shawn Hervey-Jumper, MD: No financial relationships to disclose

Introduction: Checkpoint inhibitor (CPI) therapy has demonstrated efficacy for patients with brain metastases. However, few biomarkers exist for predicting CPI treatment response. In this study, we examined the impact of tumor mutational burden (TMB) on CNS progression events for patients who underwent resection of a brain metastasis with perioperative CPI treatment.

Methods: A retrospective, single-center study was conducted on 57 patients who underwent resection of brain metastases and received CPI treatment. TMB data was obtained from next generation sequencing of brain metastasis specimens. Local and distant CNS progression as well as postoperative leptomeningeal disease (LMD) diagnosis were outcomes of interest. Optimal TMB cut-offs for predicting outcomes of interest were determined using recursive partitioning analysis and iterative threshold search. The Kaplan-Meier method and Cox proportional hazard analyses were performed to assess TMB’s association with CNS progression.

Results: The cohort consisted of 57 patients with various cancer types with the three most common being melanoma (n=27), NSCLC (n=11), and breast cancer (n=4). Median follow-up was 7 months. At last follow-up, 8 patients (14%) had local progression, 27 (47%) had distant progression, and 10 (18%) developed LMD. Median tumor mutational burden (TMB) was 14 mutations/Mb. Patients with lower TMB experienced faster local progression (High vs Low TMB: median not reached for both, p=0.004) and decreased time to LMD diagnosis (High vs Low TMB: median not reached vs 12.5 months, p=0.049). Patients with lower TMB also experienced a trend for faster distant progression (High vs Low TMB: 9.77 vs 5.53 months, p=0.087). In a multivariate Cox proportional hazards model adjusting for age, sex, preoperative KPS, primary cancer, extent of resection, and postoperative RT, lower TMB predicted faster local recurrence (log(HR) 11 [3.4, 18], p=0.004), distant progression (log(HR) 1 [0.17, 1.9], p=0.018), and faster LMD (log(HR) 2.6 [0.13, 5.1], p=0.039).

Conclusion : Low tumor mutational burden is associated with decreased time to local and distant progression, and LMD for patients who undergo brain metastasis resection with perioperative CPI treatment.