Empirical Relative Biological Effectiveness (RBE) for Mandible Osteoradionecrosis (ORN) in Head and Neck Cancer Patients Treated With Pencil-Beam-Scanning Proton Therapy (PBSPT): A Retrospective, Case-Matched Cohort Study

Abstract

PurposeTo retrospectively investigate empirical relative biological effectiveness (RBE) for mandible osteoradionecrosis (ORN) in head and neck (H&N) cancer patients treated with pencil-beam-scanning proton therapy (PBSPT).MethodsWe included 1,266 H&N cancer patients, of which, 931 patients were treated with volumetric-modulated arc therapy (VMAT) and 335 were treated with PBSPT. Among them, 26 VMAT and 9 PBSPT patients experienced mandible ORN (ORN group), while all others were included in the control group. To minimize the impact of the possible imbalance in clinical factors between VMAT and PBSPT patients in the dosimetric comparison between these two modalities and the resulting RBE quantification, we formed a 1:1 case-matched patient cohort (335 VMAT patients and 335 PBSPT patients including both the ORN and control groups) using the greedy nearest neighbor matching of propensity scores. Mandible dosimetric metrics were extracted from the case-matched patient cohort and statistically tested to evaluate the association with mandibular ORN to derive dose volume constraints (DVCs) for VMAT and PBSPT, respectively. We sought the equivalent constraint doses for VMAT so that the critical volumes of VMAT were equal to those of PBSPT at different physical doses. Empirical RBEs of PBSPT for ORN were obtained by calculating the ratio between the derived equivalent constraint doses and physical doses of PBSPT. Bootstrapping was further used to get the confidence intervals.ResultsClinical variables of age, gender, tumor stage, prescription dose, chemotherapy, hypertension or diabetes, dental extraction, smoking history, or current smoker were not statistically related to the incidence of ORN in the overall patient cohort. Smoking history was found to be significantly associated with the ORN incidence in PBSPT patients only. V40Gy[RBE], V50Gy[RBE], and V60Gy[RBE] were statistically different (p<0.05) between the ORN and control group for VMAT and PBSPT. Empirical RBEs of 1.58(95%CI: 1.34-1.64), 1.34(95%CI: 1.23-1.40), and 1.24(95%: 1.15-1.26) were obtained for proton dose at 40 Gy[RBE=1.1], 50 Gy[RBE=1.1] and 60 Gy[RBE=1.1], respectively.ConclusionsOur study suggested that RBEs were larger than 1.1 at moderate doses (between 40 and 60 Gy[RBE=1.1]) with high LET for mandible ORN. RBEs are underestimated in current clinical practice in PBSPT. The derived DVCs can be used for PBSPT plan evaluation and optimization to minimize the incidence rate of mandible ORN.

Exploratory study of seed spots analysis to characterize dose and linear‐energy‐transfer effect in adverse event initialization of pencil‐beam‐scanning proton therapy

“…This dose engine has been fully commissioned and benchmarked with other MC codes 68,71–75 . It has been incorporated in our in‐house treatment planning system, Shiva, 39,55,76–79 and clinically used as our second monitor unit check system 68,71 …”

“…[67][68][69][70] This dose engine has been fully commissioned and benchmarked with other MC codes. 68,[71][72][73][74][75] It has been incorporated in our in-house treatment planning system, Shiva, 39,55,[76][77][78][79] and clinically used as our second monitor unit check system. 68,71 The dose-averaged LET 45,69 was used to calculate the LET at each voxel from multiple fields.…”

“…The appropriateness of a fixed RBE value of 1.1 has come under scrutiny. Both in vitro cell experiments 43,47,48 and clinical studies 49–55 indicated a monotonic increase of RBE along the beam direction. Ignoring the effects of high LET distributions may lead to unexpected AEs in cancer patients treated by PBS.…”

Technical note: Investigation of dose and LETd effect to rectum and bladder by using non‐straight laterals in prostate cancer receiving proton therapy

“…The adverse events after PBT are more complex than photon therapy due to the unique non-linear characteristic of the relationship of RBE with dose and LET d in PBT. 6,8,[10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27] A newly developed tool, dose-LET d volume histogram (DLVH), was developed and was able to correlate high LET d and high dose with rectal adverse events. 26,28 It was reported that high LET d (≥2.5 keV/µm) volume in the rectum might result in a higher incidence rate of grade≥2 (CACTE V4.0) rectal bleeding in prostate cancer patients treated with proton therapy, even though the corresponding dose volume was moderate.…”

Achievements and challenges in normal tissue response modelling for proton therapy

“…In addition, Yang et al recently estimated empirical RBEs for mandible osteoradionecrosis in HNC patients treated with pencil beam scanning protons versus photon-based volumetric modulated arc therapy, using a case-matched cohort design. Depending on dose cut-off, the estimated RBEs were 1.24–1.58 at moderate dose levels [29] . Evidently, there is a need for conventions and development of appropriate methods for clinical studies of proton RBE, and to harmonize LET collection across centers [30] .…”