Histopathology-derived modeling of prostate cancer tumor control probability: Implications for the dose to the tumor and the gland.

Abstract

MATERIAL AND METHODS

Tumors' cell numbers (N0) and Gleason Scores (GS) were derived from histopathology of 25 specimens. Index lesions and tumors ⩾0.5cm(3) were considered GTV. Satellites <0.5cm(3) constituted the tumor load in the CTV. Each patient's tumor control probability (TCP) was simulated using the linear quadratic model and considering the N0 while assuming either a constant or GS-dependent α and β.

RESULTS

19/25 patients had multi-focal disease. In 11 patients the CTV contained GS 4+3 or 4+4 tumors. Compared to the GTV, the CTV pathology was more favorable. For an α=0.140Gy(-1), a GTV dose of 79Gy with a CTV dose of 72Gy achieved an 80% TCP in the population. Varying α between 0.160-0.118Gy(-1) with GS, a GTV and CTV dose of 80Gy and 70Gy also gave an 80% TCP.

CONCLUSIONS

Considering only N0, our simulations suggest that a GTV-CTV dose differentiation of 7Gy would not compromise TCP of the patient population. When assuming an increased radiosensitivity with lower GS, a further dose differentiation of 10Gy might be feasible.

PURPOSE

To evaluate the impact of GTV-CTV dose differentiation by simulating response of prostate patients to radiotherapy, considering histopathology of prostatectomy specimens.