Comparison of Oncologic Outcomes and Treatment-Related Toxicity of Carbon Ion Radiotherapy and En Bloc Resection for Sacral Chordoma

Key Points Question How do the outcomes of carbon ion radiotherapy (CIRT) compare with the outcomes obtained with en bloc surgical resection for sacral chordoma? Findings In this cohort study including 911 patients with sacral chordomas, CIRT provided similar tumor control and survival outcomes compared with en bloc surgery but with a lower rate of peripheral motor neuropathy. Meaning These findings suggest that CIRT is useful treatment for older patients with high performance status and sacral chordoma in whom surgery is not preferred.


. Kaplan-Meier plots comparing OS between matched ¥ CIRT and primary radiotherapy NCDB cohorts
eFigure 5A displays patients treated with primary photon radiotherapy (RT); eFigure 5B displays patients treated with primary proton radiotherapy (RT). ¥ 1:1 matching was performed for age and sex in eFigure 5A. 3:1 matching was performed for age and sex in eFigure 5B.

eTable 5. Radiotherapy treatment factors by modality for unmatched NCDB patients who underwent primary radiotherapy (RT) and for primary CIRT* patients CIRT (n=188) Photon RT (n=80) § Proton RT (n=30) ¶ P-value
Age, median However, patients in the NCDB cohort treated with photon therapy alone were treated to a relatively low dose which prevented an adequate comparison to protons and CIRT. In the small group of proton patients available in our study, the direct comparison to CIRT did not reveal any significant difference in overall survival recognizing, however, that the biologically effective dose (BED) used for the CIRT patients was significantly higher than the proton patients. Despite the small number of patients, the outcomes for CIRT should theoretically be superior to proton radiotherapy due to the higher linear energy transfer (LET), lower oxygen enhancement ratio (OER), and higher relative biological effectiveness (RBE). However, when controlling for BED, there was still no difference in OS.
Overall survival may not be a good surrogate for local tumor control because there may be a difference in local tumor control which is not reflected in OS. In contrast, patients undergoing radiotherapy alone with photons had a lower overall survival compared to the CIRT group (p<0.001), even when controlling for BED (p<0.001) suggesting a possible LET, OER, and RBE advantage to CIRT. The data from this study is being used to design a prospective trial sponsored by Mayo Clinic comparing outcomes for patients treated with CIRT or proton therapy.
It is also theorized that these potential advantages of CIRT may be overcome through hypofractionation of protons.
There is an ongoing Phase I/II trial comparing proton radiotherapy to CIRT in a population of patients with evidence of macroscopic tumor [Ion Irradiation of Sacrococcygeal Chordoma (ISAC), NCT01811394]. 1 This study allows patients treated with both radiotherapy alone and margin-positive surgery with adjuvant radiotherapy. However, separate evaluations for the two groups would be a better comparison. In addition, there are concerns that the trial design may not allow adequate evaluation of the outcomes due to the short 3-year tumor control and survival endpoint, the relatively low CIRT dose of 64 Gy (RBE3.0) and using different methods of estimating RBE in the CIRT treatment planning systems.

eAppendix 2. Cost Analysis
Two-year standardized Medicare costs were available for 28 patients treated with en bloc resection and 15 patients treated with proton radiotherapy at Mayo Clinic (eTable 6). Both mean total costs and total procedural costs were significantly higher for the en bloc resection cohort at $137,182 (SD $117,894) and $110,375 (SD $94,169) versus $68,066 (SD $60,567) and $53,699 (SD $43,052) for the proton group, p=0.02 and p<0.01 respectively. Proton radiotherapy costs at Mayo Clinic of $30,489 (SD $4,530) were comparable to CIRT costs at QST Hospital of $22,800. This equates to an estimated price per fraction of $924 for proton radiotherapy (33 fraction course) and $1,425 for CIRT (16 fraction course). 2 An estimate for possible U.S. Medicare reimbursement per fraction of CIRT was calculated to be $1,906 per fraction assuming price parity between courses of proton radiotherapy and CIRT treatment episodes.
Despite the small numbers of patients treated comparatively with proton radiotherapy, OS was similar to patients treated with CIRT, and both were superior to primary photon radiotherapy. Our cost analysis shows that CIRT total and procedural costs in the U.S. are likely to be on par with that of proton radiotherapy. While the theoretical benefits of LET, OER, and RBE of CIRT may result in improved local control, although perhaps not OS, ongoing and future research studies are vital to understanding these differences and improving radiotherapy modalities.