[18F]FluorThanatrace PET imaging as a biomarker of response to PARP inhibitors in breast cancer

Abstract

Background: Poly (ADP-ribose) polymerase inhibitors (PARPi) are approved for Breast Cancer gene (BRCA)-mutant HER2- breast cancer, and there is clinical interest in expanding indications to include homologous recombination deficient (HRD) breast cancers. Yet, response in these populations remains variable, suggesting clinical utility in developing a better biomarker to select patients for PARPi and predict response. Here, we evaluate a radiolabeled PARPi, [¹⁸F]FluorThanatrace ([¹⁸F]FTT), as a functional biomarker of PARPi response in breast cancer.

Methods: A single-arm prospective observational trial was conducted at the University of Pennsylvania. [¹⁸F]FTT-PET uptake was measured in 24 women with untreated primary breast cancer and correlated with tumor HRD score. In a separate cohort of ten subjects with metastatic HER- breast cancer, [¹⁸F]FTT-PET uptake was measured at baseline and after a short interval on a PARPi (a measure of drug-target engagement) and correlated to progression free survival (PFS).

Results: Here we show that baseline [¹⁸F]FTT-PET uptake does not correlate to HRD tissue score, supporting that [¹⁸F]FTT provides distinct information from genetic features. Baseline [¹⁸F]FTT-PET uptake and the change in uptake from baseline to after PARPi initiation significantly correlates to PFS in woman with breast cancer who received a PARPi (ρ = 0.74, P = 0.023 and ρ = -0.86, P = 0.012, respectively).

Conclusions: These early results suggest the potential of [¹⁸F]FTT-PET to select patients for PARPi treatment and monitor in vivo pharmacodynamics after therapy start. Absence of association with HRD scores supports [¹⁸F]FTT uptake as a novel measure that may be leveraged as a biomarker. Further studies are warranted.

Fig. 1. HRD does not correlate the [¹⁸F]FTT uptake.

Baseline [¹⁸F]FTT uptake was measured in the primary breast lesion of 24 subjects with breast cancer prior to standard of care therapy. Pearson correlation of [¹⁸F]FTT maximum standardized uptake values (SUV_max) at baseline versus tumor homologous recombination deficiency (HRD) score. Each data points represent one subject (n = 24). Linear regression of mean SUV_max is plotted.

Fig. 2. [¹⁸F]FTT uptake is broad and does not correlate to lesion size.

A Schematic representation of [¹⁸F]FTT-PET imaging trial. 10 women received a baseline [¹⁸F]FTT-PET/CT scan, and 8 women received a second optional [¹⁸F]FTT-PET/CT scan after PARPi therapy initiation. Tumors were monitored until progression (n = 8 subjects). B Distribution of [¹⁸F]FTT uptake (maximum standardized uptake values, SUV_max) in subjects (n = 10). Each point represents a target lesion (n = 22). The mean and SD for each patient is shown. C Pearson correlation of tumor diameter (cm, centimeter) measured by CT imaging and baseline [¹⁸F]FTT uptake per lesion. Linear regression is plotted.

Fig. 3. Representative [¹⁸F]FTT-PET imaging in a PARPi responder and non-responder.

Representative clinical imaging of [¹⁸F]FTT-PET/CT in a PARPi responder and non-responder. PARPi-responder (left; germline PALB2) pre-PARPi baseline imaging of ER + PR + HER- lobular breast cancer demonstrated [¹⁸F]FTT uptake in a metastatic left axillary soft tissue mass marked with a metallic biopsy clip (arrow) (SUV_max of 7.0). Imaging after starting olaparib demonstrates decreased [¹⁸F]FTT uptake (SUV_max of 2.7). Non-responder (right; germline PALB2) pre-PARPi imaging of TNBC demonstrates low uptake in the index lung metastasis (SUV_max of 3.2, arrow) suggestive of low PARP-1 expression and PARPi drug target with decreased uptake after starting olaparib (SUV_max of 2.1).

Fig. 4. [¹⁸F]FTT uptake predicts response to PARPi therapy.

A Spearman correlation of [¹⁸F]FTT maximum standardized uptake values (SUV_max) at baseline versus progression free survival (PFS). Data points represent the average of [¹⁸F]FTT uptake across lesions per subject and error bars represent the range of [¹⁸F]FTT uptake per subject (n = 8). Linear regression of mean SUV_max is plotted. B Baseline [¹⁸F]FTT SUV_max in lesions from subjects who had a >5 month (mo) PFS compared to those who had a <5 month PFS (n = 8). Data points represent the average of [¹⁸F]FTT uptake across lesions per subject and median uptake. Mann–Whitney test, *P = 0.038. C Change in [¹⁸F]FTT uptake from baseline to after PARPi initiation per lesion. (n = 16 lesions in 8 subjects). Data is shown as individual target lesions. Paired non-parametric T test, ****P = 0.00003. D Spearman correlation of the percent change in SUV_max from baseline to post-PARPi therapy initiation versus PFS (n = 7). Data points represent the average of [¹⁸F]FTT uptake across lesions per subject and error bars represent the range of [¹⁸F]FTT uptake per subject. Linear regression of mean SUV_max is plotted.

Fig. 5. Comparative [¹⁸F]FDG and [¹⁸F]FTT imaging.

71 year old female with ER/PR + invasive ductal breast cancer with a known somatic BRCA2 mutation. [¹⁸F]fluorodeoxyglucose(FDG)-PET/CT (PET (left), CT (middle), PET/CT (right)) imaging demonstrate an FDG-avid mass in the caudate lobe of the liver with [¹⁸F]FDG uptake throughout the mass suggesting tumor viability. [¹⁸F]FTT-PET/CT demonstrates a photopenic center, suggesting either decreased PARP-1 expression centrally or decreased drug delivery to the center aspect.