# Variable cellular and radiobiological effects of [177Lu]Lu-PSMA-I&T in patient-derived models of prostate cancer

## TL;DR

This study explores why some prostate cancer patients respond well to [177Lu]Lu-PSMA therapy while others do not, using patient-derived tumor models.

## Contribution

The study identifies intrinsic tumor characteristics beyond PSMA expression that influence response to [177Lu]Lu-PSMA therapy.

## Key findings

- Tumor growth response to [177Lu]Lu-PSMA is not reliably predicted by PSMA expression level or radioligand uptake.
- Divergent DNA damage response and cell death occur following [177Lu]Lu-PSMA treatment due to intrinsic tumor characteristics.
- Baseline tumor transcriptomic profiles may serve as predictors of response to [177Lu]Lu-PSMA therapy.

## Abstract

[177Lu]Lu-PSMA radioligand therapy targets metastatic castration-resistant prostate cancer by delivering radiation to cells expressing prostate-specific membrane antigen (PSMA). While some patients show remarkable responses, up to 50% show little to no benefit, and disease progression inevitably occurs. Eligibility is mainly based on PSMA avidity by PET imaging, yet responses remain highly variable, highlighting a disconnect between PSMA expression and clinical efficacy. This study used PSMA-positive patient-derived xenografts (PDXs) from the Melbourne Urological Research Alliance (MURAL) to explore mechanisms of response and resistance to [177Lu]Lu-PSMA treatment.

PDXs with variable PSMA expression were treated with [177Lu]Lu-PSMA-I&T and assessed for radioligand uptake and DNA damage. Histological, genomic and transcriptomic analyses aimed to identify features of radiosensitivity and resistance at a tissue and cellular level.

PDXs recapitulated clinical PSMA histology and exhibit variable responses to [177Lu]Lu-PSMA, mirroring the heterogeneity observed in the clinic. Responses ranged from sustained, transient or delayed tumor reduction to treatment resistant, including a remarkable responder with complete tumor regression for up to 20 weeks post treatment. The magnitude and persistence of DNA damage and tumor cell death varied between PDXs, demonstrating divergent radiosensitivity due to intrinsic tumor characteristics. These characteristics included PSMA receptor density, extent of DNA damage, genomic aberrations and baseline transcriptomic signatures.

These findings in PDX tumors emphasize the complexity of predicting [177Lu]Lu-PSMA response, beyond what cell line models can capture. Comparable radiation doses with variable regression underscores that radiosensitivity is multifactorial across diverse tumors, supporting the need for integrated multimodal approaches for patient stratification. Integrating baseline genomic and transcriptomic profiles may reveal determinants of [177Lu]Lu-PSMA sensitivity to improve patient selection and identify novel combination therapies.

The online version contains supplementary material available at 10.1186/s13046-026-03659-w.

Patient-derived prostate cancer models capture PSMA heterogeneity, providing a robust platform for preclinical theranostic studies.Tumor growth response to [177Lu]Lu-PSMA is not reliably predicted by PSMA expression level or radioligand uptake.Divergent DNA damage response and cell death occurs following [177Lu]Lu-PSMA treatment due to intrinsic tumor characteristics.Baseline tumor transcriptomic profiles may serve as predictors of response to [177Lu]Lu-PSMA therapy.

Patient-derived prostate cancer models capture PSMA heterogeneity, providing a robust platform for preclinical theranostic studies.

Tumor growth response to [177Lu]Lu-PSMA is not reliably predicted by PSMA expression level or radioligand uptake.

Divergent DNA damage response and cell death occurs following [177Lu]Lu-PSMA treatment due to intrinsic tumor characteristics.

Baseline tumor transcriptomic profiles may serve as predictors of response to [177Lu]Lu-PSMA therapy.

The online version contains supplementary material available at 10.1186/s13046-026-03659-w.