Pencil Beam Scanning Proton Therapy In The Treatment Of Rectal Cancer

Radiation therapy is standard approach in the neoadjuvant setting for the treatment of locally advanced rectal cancer. New technological advances may allow for proton therapy to be integrated in the neoadjuvant treatment of rectal cancer. This study evaluates the dosimetric and clinical results of the use of pencil beam scanning proton therapy (PBSPT) in the treatment of locally advanced rectal cancer (LARC). Neoadjuvant chemoradiation therapy utilizing PBSPT (50.4 Gy RBE to standard volumes with concomitant 5-FU or capecitabine) was administered to ≥ Stage II LARC patients on a prospective registry and quality of life study. PBSPT was delivered using two opposed lateral fields in a fixed beam line with the patient positioned supine with a comfortably full bladder. A proton planning target volume was created to account for range uncertainty in the beam direction. Daily pretreatment verification was performed with orthogonal kilovoltage images registered to bony anatomy and weekly offline CT scans were performed for plan robustness analysis. Back-up photon treatment plans were created using volumetric arc therapy or 8-field IMRT. For PBSPT vs photon plan comparison, conformity index (CI), homogeneity index, main organ at risk (OAR) sparing and the integral dose (ID) were calculated. Statistical analysis was performed using a pairwise Wilcoxon signed-rank test. Between November 2012 and February 2013, five patients were treated with PBSPT; no >G2 toxicities were observed. CI values were high for both proton and photon plans (1.04 ± 0.15 vs 1.05 ± 0.14, respectively, p = NS). PBSPT plans were slightly more homogenous (0.06 ± 0.01 vs 0.1 ± 0.02, p = NS). Target coverage was up to 95% of the prescription dose for more than 99% of target volume for both. Analysis of robustness revealed consistent target coverage (within 2% of original plan) in all verification plans. The biggest source of uncertainty along the beam path was rectal and bowel gas. Bladder filling varied an average of 67 cc on the verification scans. Bowel average V15 was 65.8 cc with PBST vs 286.2 cc with IMRT (p = 0.04); bladder D50 was 14.8 Gy vs 25.9 Gy (p = 0.04). Femoral head D50 was 12.9 Gy vs 19 Gy (p = 0.01). Average ID was 111.5 J vs 252.5 J (p = 0.04). These preliminary results indicate a potential role of proton therapy in the treatment of LARC, a disease with high cure rates where radiation related toxicities should be minimized. There have been no acute toxicity >G2 in the initial cohort of patients. A prospective multimodality protocol is under development.