Pre-treatment tumour PET metrics and clinical outcomes of anal cancer in patients living with and without HIV.

Abstract

Background/purpose: To investigate if pre-treatment tumour positron-emission tomography (PET) metrics' prognostic efficacy changes with HIV or viral load (VL) in anal squamous cell carcinoma (ASCC).

Materials and methods: Consecutive patients treated with definitive radiation therapy (RT) for non-metastatic ASCC from 2005 to 2021 at one institution were retrospectively identified. Patient demographic and clinical data, including HIV status and pre-treatment VL, were tabulated. Pre-treatment PET metrics were calculated with semi-automatic gradient-based segmentation algorithms. Cox-proportional-hazard and Kaplan-Meier modelling were used to investigate tumour PET metrics and outcomes: overall survival (OS), progression-free survival (PFS), and locoregional control (LRC).

Results: A total of 175 patients were included: 110 HIV-negative and 65 patients living with HIV (PLWH). Nineteen PLWH had detectable pre-treatment VL. Median follow-up was 58 months (interquartile range [IQR]: 28-99), with 28 locoregional failures and 31 deaths. Five-year LRC, PFS, and OS was 84%, 73%, and 86%, respectively. There was no significant difference in LRC, PFS, or OS between HIV-negative patients and PLWH. 156 patients had available pre-treatment PET scans. Metabolic tumour volume and total lesion glycolysis were significantly associated with LRC and PFS on multivariate Cox analysis for the entire cohort (p ≤ 0.02), and HIV-negative patients on Cox sub-group analysis (p ≤ 0.01). No association between PET metrics and outcomes was seen for PLWH.

Interpretation: Outcomes were comparable between HIV-negative patients and PLWH. Pre-treatment PET metrics were validated as significantly predicting outcomes for the entire cohort and HIV-negative patients, not PLWH. This may be from small numbers of PLWH patients, or non-specific uptake in patients with uncontrolled HIV reducing PET's prognostic efficacy.