Quantification of renal function following stereotactic body radiotherapy for pancreatic cancer: secondary dosimetric analysis of a prospective clinical trial

BACKGROUND: This is the first known study examining renal function following stereotactic body radiotherapy (SBRT) for pancreatic head adenocarcinoma. METHODS: Thirty-eight borderline-resectable/unresectable patients, part of an ongoing prospective trial, underwent 3 cycles of gemcitabine/5-fluorouracil followed by SBRT (5 daily fractions of 5/6/7/8 Gy) and concurrent nelfinavir. Thereafter, in resectable cases, surgery was performed within 4–8 weeks. The last available pre-SBRT creatinine was recorded, along with the highest post-SBRT value. Glomerular filtration rate (GFR) was calculated by the commonly-utilized Modification of Diet in Renal Disease formula. GFR decline was defined as the post-SBRT nadir GFR minus the pre-SBRT GFR. Correlations with the V5–V30, and mean/maximum kidney doses was performed. Statistics included Pearson correlation, Mann-Whitney, and Fisher’s exact tests. RESULTS: The median total kidney volume was 355 cm(3). Median dosimetric values were as follows: V5 (209 cm(3)), V10 (103 cm(3)), V15 (9 cm(3)), V20 (0 cm(3)), V25 (0 cm(3)); and mean (6.7 Gy) & maximum kidney dose (18.3 Gy). Median GFR change was −23 (range, −105 to 25) mL/min/1.73 cm(2). Of all dosimetric parameters, only V5 was significantly associated with changes in GFR (Pearson r = −0.40, p = 0.012). In patients with V5 < 210 cm(3), median GFR change was −11.8 mL/min/1.73 cm(2), as compared with −37.1 mL/min/1.73 cm(2) change in those with V5 ≥ 210 cm(3) (p = 0.02). A GFR change < −23 mL/min/1.73 cm(2) was observed in 6/20 (30%) patients with V5 < 210 cm(3), versus 15/18 (83%) of those with V5 ≥ 210 cm(3). Patients with V5 ≥ 210 cm(3) were over ten times as likely to have GFR change < −23 mL/min/1.73 cm(2) (p = 0.003). Using linear regression, GFR change ≈ −0.1748 × V5(cm(3)) + 8.63. CONCLUSIONS: In the first known analysis of renal function after pancreatic SBRT, evaluating patients on a prospective study, V5 ≥ 210 cm(3) was associated with a post-SBRT GFR decline of >23 mL/min/1.73 cm(2). If V5 is kept <210 cm(3), median GFR decline was only 11.8 mL/min/1.73 cm(2). Further validation is needed to ascertain definite dose-volume parameters and examine late renal decline.