Purpose/Objective(s)

Hypofractionated external beam radiotherapy with 10 to 30 fractions (fx) and stereotactic body radiation therapy (SBRT) for prostate cancer (PC) have similar rates of cancer control and toxicity at 5 years as compared to conventional treatment. There is a lack of a radiobiological modeling from lab experiments to guide this transition. By assuming a unified multi-activation (UMA) model of lab-measured radiosensitivity (RS) of the cell lines to the entire dose range, we explored the clinical dose-response, modeling tumor-control probability (TCP) and normal-tissue complication probability (NTCP).

Materials/Methods

Published cell survival curves (CSC) with experimental error bars for PC of 5 DU145 CSCs, 3 CP3 CSCs, 6 LNCaP + 2 LnCaP with BCL2 overexpressed or down-regulated CSCs, 3 wild-type (WT)-22RV1 CSCs, 3 post-radiation radioresistant (RR)-22RV1 CSCs, 2 hypoxia-22RV1 CSCs, 3 urothelial HCV-29 CSCs and 4 large intestine (LI) or small intestine (SI) stem cell and organoid CSCs were fitted with a UMA model by using a logit transformation and c²-regression to obtain the model parameters of g and n and their deviations. Cell survival at any fractional dose (=total dose/fx) and TCP and NTCP for the total number of cancer cells or normal tissue cells corresponding to a grade of toxicity were analytically determined from the parameters.

Results

All of the CSCs had been excellently fitted with R² > 0.95 and Q-factor > 0.2 excepting for two in-vivo LI/SI crypt CSCs each with an outline point. We identified radiobiological parameters and predicted TCP/NTCP curves and listed the data for the 8Gyx5 SBRT of PC in Table 1 that are comparable with the results listed in the HyTEC papers. The 0% TCP to RR or hypoxia PC and 100% NTCP to SI are similar to conventional EBRT. The Dg reflects the major uncertainties in the lab-measured RS. D50, the total dose at 50% TCP or NTCP, correlated with RS and number of cells. Cell heterogeneity greatly changed RS and shapes of TCP and NTCP curves.

Conclusion

These findings may be useful for the design and evaluation of new clinical treatment schemes in PC translating lab work into clinical application.

Dose-Response Prediction for SBRT of Prostate Cancer Based on Measured Cell Survival Curves Using a Unified Multi-Activation Model