Independent validation of a dysphagia dose response model for the selection of head and neck cancer patients to proton therapy

Impact of Modern Low Dose Involved Site Radiation Therapy on Normal Tissue Toxicity in Cervicothoracic Non-Hodgkin Lymphomas: A Biophysical Study

“…Another factor that influences the quality of data included in the NTCP model adaption might be the interobserver variability among different clinicians for head and neck cancer patients. As reported by Kalendralis and Sloep et al, these variabilities can directly influence the performance of the adapted NTCP model in different independent patient cohorts [44]. It is also worth noting that in a related study, the biophysical model displayed sensitivity to parameter variations, where even a 1% deviation in TD 50 resulted in a substantial 10% deviation in the NTCP [6,23].…”

“…Multiple studies have evaluated the reduction in treatment volume and prescribed dose as well as the adoption of modern conformal techniques, in order to limit treatment-induced side-effects [2]. In 2011, a large, randomized trial confirmed the equivalence in terms of response rate and local control of 24 Gy for indolent NHL and 30 Gy in 2 Gy fractions for aggressive NHL, compared with the doses previously considered as standard (40)(41)(42)(43)(44)(45) [3]. Furthermore, the reduction in treatment volume from mantle field to involved-field radiotherapy first and further to involved-site or -node radiotherapy demonstrated comparable control of disease with reduced irradiation of organs at risk (OAR) [4].…”

Achievements and challenges in normal tissue response modelling for proton therapy

“…Both studies applied the closed testing procedure [14] , which is an automated method that evaluate to what degree an NTCP model needs to be adjusted in order to improve performance when applied in the new patient cohort. The results from the external validation in the above mentioned studies differed, in the Danish patient cohort, updating the intercept resulted in a significant improvement in the model performance, whereas re-estimation of model coefficients significantly improved the performance in Kalendralis et al [13] .…”

“…In summary, the contribution of Kalendralis et al [13] in the current volume is an important achievement within a research area that is essential for the further expansion of the scientifically founded indications of proton therapy.…”

“…Recently, a previously published NTCP model for dysphagia [11] was externally validated in a Danish patient cohort [12] and in the current volume of this journal, Kalendralis et al [13] report the results from an external validation of an NTCP model for grade II dysphagia in 270 HNC patients. Both studies applied the closed testing procedure [14] , which is an automated method that evaluate to what degree an NTCP model needs to be adjusted in order to improve performance when applied in the new patient cohort.…”