Can We Predict Deep PSA Response Before Treatment Starts? NADIR Validated in ARCHES

A new analysis of the phase 3 ARCHES trial suggests that a simple clinical model may help predict which men with metastatic hormone-sensitive prostate cancer are most likely to achieve a very deep PSA response after starting treatment with an androgen-receptor pathway inhibitor.
The model, called NADIR, was designed to estimate the probability of reaching a PSA of 0.2 ng/mL or lower within six months. It uses routine baseline information such as PSA, haemoglobin, disease volume and the presence of visceral metastases, rather than requiring a new molecular test.
In the ARCHES validation, 558 patients treated with enzalutamide had sufficient data for analysis, and 359, or 64%, reached PSA ≤0.2 ng/mL within six months. The model showed good predictive performance, with an AUC of 0.80.

The separation between risk groups was substantial. About 92% of patients in the highest predicted-response group reached PSA ≤0.2, compared with 64% in the intermediate group and 37% in the lowest group. The predicted probability of a deep PSA response was also associated with overall survival, suggesting that NADIR may capture more than PSA behaviour alone.
The potential value is that NADIR could identify, very early, patients who are unlikely to achieve a deep response with standard AR-pathway treatment. In the future, such patients might be candidates for earlier treatment intensification or alternative strategies rather than waiting for obvious progression.

There are important caveats. NADIR is currently a prognostic tool, not a treatment-selection test, and prospective studies are needed before it can be used to decide whether therapy should be changed. A low predicted probability does not prove that a patient will do poorly, and a deep PSA response does not guarantee that the disease is fully controlled.
This last point is especially important because ARCHES has also shown that some patients can develop radiographic progression without a clear PSA rise. PSA therefore remains extremely useful, but it cannot completely replace imaging.

Source.