The value of individual measurements for tumor control probability predictions in head and neck patients

• In the age of personalized cancer medicine, individual measurements of in vitro radiosensitivity and proliferation parameters have great potential for predicting treatment outcome. However, cellular radiosensitivity is quite heterogeneous and therefore concerns exist towards its impact on treatment predictions. It was therefore the purpose of this study to investigate this aspect. Individually-determined radiosensitivities and potential doubling times, as well as tumor volumes from 46 head-and-neck carcinomas treated with radiotherapy, were used to predict tumor control probabilities (TCP) under various biologically-relevant assumptions for heterogeneity in radiosensitivity. TCP predictions were then compared to clinical local control using a ROC curve analysis. The analysis showed that TCP calculated under the assumption of heterogeneous radiosensitivity have the same power of distinguishing between patients with or without local control as from single values for the radiobiological parameters (a sensitivity of 66% and a specificity of 80% for an area under the curve of 0.69). The only difference was in the discrimination criterion (TCP>93% for single parameters and TCP>65% for heterogeneous parameters), illustrating the difference in appearance of the TCP curve under the assumption of heterogeneity. Nevertheless, the results showed that individually determined radiobiological parameters could be quite effective towards predicting treatment outcome for individual patients.

Ämnesord

MEDICIN OCH HÄLSOVETENSKAP  -- Klinisk medicin -- Cancer och onkologi (hsv//swe)

MEDICAL AND HEALTH SCIENCES  -- Clinical Medicine -- Cancer and Oncology (hsv//eng)

Nyckelord

radiosensitivity

potential doubling time

head-and-neck carcinomas

TCP

Publikations- och innehållstyp

ref (ämneskategori)

kon (ämneskategori)