Therapeutic Efficacy of Variable Biological Effectiveness of Proton Therapy in U-CH2 and MUG-Chor1 Human Chordoma Cell Death

Abstract

Background: Chordoma is a cancer of spinal cord, skull base, and sacral area. Currently, the standard of care to treat chordoma is resection followed by radiation therapy. Since, chordoma is present in the spinal cord and these are very sensitive structures and often complete removal by surgery is not possible. As a result, chordoma has a high chance of recurrence and developing resistance to radiation therapy. In addition, treatment of chordoma by conventional radiation therapy can also damage normal tissues surrounding chordoma. Thus, current therapeutic options to treat chordoma are insufficient and novel therapies are desperately needed to treat locally advanced and metastatic chordoma. (2) Methods: In the present investigation, human chordoma cell lines of sacral origin MUG-Chor1 and U-CH2 were cultured and irradiated with Proton Beam Radiation using the clinical superconducting cyclotron and pencil-beam (active) scanning at Middle and End of the Spread-Out Bragg Peak (SOBP). Proton radiation was given at the following doses: Mug-Chor1 at 0, 1, 2, 4, and 8 Gy and U-CH2 at 0, 4, 8, 12, and 16 Gy. These doses were selected based on a pilot study in our lab and attempted to produce approximate survival fractions in the range of 1, 0.9, 0.5, 0.1, and 0.01, respectively, chosen for linear quadratic model fitting of the dose response. (3) Results: In this study, we investigated relative biological effectiveness (RBE) of proton radiation at the end of Spread Out Bragg Peak assuming that the reference radiation is a proton radiation in the middle of the SOBP. We observed differences in the survival of both Human chordoma cell lines, U-CH2 and MUG-Chor1. The data showed that there was a significantly higher cell death at the end of the Bragg peak as compared to middle of the Bragg peak. Based on the linear quadratic (LQ) fit for cell survival we calculated the RBE between M-SOBP and E-SOBP at 95% CI level and it was observed that RBE was higher than 1 at E-SOBP and caused significantly higher cell killing. Proton field at E-SOBP caused complex DNA damage in comparison to M-EOBP and the genes such as DNA topoisomerase 1, GTSE1, RAD51B were downregulated in E-SOBP treated cells. Thus, we conclude that there seems to be substantial variation in RBE (1.3–1.7) at the E-SOBP compared with the M-SOBP.

Impact of proton therapy on the DNA damage induction and repair in hematopoietic stem and progenitor cells

“…The fact that no change in the number of γ-H2AX foci was observed at 30 min and 24 h, while a higher number of MN was observed after exposure to distal-SOBP compared to mid-SOBP protons, might be explained by a higher complexity of the DSBs induced by distal-SOBP protons. As mentioned before, this outcome has been seen in various in vitro studies and is linked to the increased LET in this region 22 – 25 .…”

“…A higher LET leads to a higher ionization density, leading to more complex DNA damage which is therefore more difficult to repair resulting in a higher yield of chromosomal aberrations. This has been seen in various in vitro studies 22 – 25 . To describe the biological effect of proton radiation, the proton relative biological effectiveness (RBE) is used.…”

“…While there is evidence for a variable RBE along the proton beam depth, this remains a topic of active debate 20 , 29 . In particular, near the distal region of the Bragg peak several in vitro and in vivo studies showed a significantly higher RBE due to the higher LET, which can be critical as this SOBP portion is likely to be located in healthy tissue 23 , 25 , 30 – 32 . These recent findings question the accuracy of a fixed RBE value along the proton beam with respect to treatment safety and efficacy.…”

Proton versus photon adjuvant radiotherapy: a multicenter comparative evaluation of recurrence following spinal chordoma resection

“…Our findings are in accordance with earlier studies suggesting the efficacy of proton therapy in managing chordomas. 15,[17][18][19] Yet, what distinguishes our study from these earlier ones is the incorporation of tree-based machine learning models, shedding light on crucial predictive variables.…”