Date of Award

January 2015

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Roy H. Decker

Subject Area(s)



Stereotactic body radiation therapy (SBRT) is a definitive treatment for limited lung cancers and pulmonary oligometastases. Our purpose was to review our institutional experience in the treatment of early stage pulmonary malignancies with SBRT to answer a number of novel questions.

Appropriate patients for individual analyses were identified in the institutional database, based on specific inclusion criteria. Toxicity was graded with the Common Terminology Criteria for Adverse Events, v4.0. Rates of local control (LC), distal control (DC), relapse-free survival (RFS), and overall survival (OS) were estimated with the Kaplan-Meier method. Additional univariate analysis (UVA) was performed with the Chi-square and two-tailed student's t-test. Multivariate analysis (MVA) was performed with Cox and logistic (LR) regressions.

The efficacy and toxicity of SBRT for stage T2 NSCLC was studied. T2 patients had significantly shorter 2-year OS on UVA (vs. T1, 38.2% vs. 59.1%, median 17.6 months vs. 31.2 months, p=0.019) and MVA (hazard ratio [HR] 1.60, 95% confidence interval [CI] 1.021-2.515, p=0.041). Higher T stage patients had non-significantly worse LC, DC, and RFS.

Similarly, the roles of diabetes mellitus (DM) and anti-diabetic medications in post-SBRT outcomes were examined. Metformin users had significantly lower rates of DC (vs. non-metformin users, 51.8% vs. 83.3%, p=0.024) and RFS (27.1% vs. 61.2%, p=0.005) on UVA and MVA. Metformin users also trended towards poorer LC and OS. LC, DC, RFS, and OS were not affected by DM or the use of insulin, sulfonylureas, thiazolidinediones, and sitagliptin.

We also studied two major methods of reducing or preventing the development of radiation pneumonitis (RP). The first analysis examined 264 primary lung cancer patients with 3+ months follow-up with a Grade 2+ RP rate of 23.1%. V10 (% volume receiving at least 10Gy) was the best predictor of RP on MVA, although V5-V45 and mean dose were also significant on UVA. In our second analysis, we studied medications that may protect against pulmonary toxicity. ACE inhibitor users experienced greater freedom from Grade 2+ RP than non-users on UVA (89.8% vs. 76.3% at 12 months, p=0.029) and MVA (HR 0.417, 95% CI 0.185-0.937, p=0.034). ARB, NSAID, steroid and statin administration were not associated with RP.

To study esophagitis dosimetric predictors, 157 patients with a planning target volume 5cm or less from the esophagus were selected. The Grade 2+ toxicity rate was 5.7%. BED10 to 1.5cc was the best predictor of Grade 2+ esophagitis on MVA (>21.1 Gy associated with 10.6% risk vs. 1.3% with >/=21.1 Gy, at 2 months). BED10 to 0.5cc, 1.0cc, 2.0-3.5cc were also predictive but less strong.

We have shown that SBRT is a safe, efficacious treatment option for both diabetic and T2 patients, although our results suggest metformin users should receive more rigorous pre- and post-SBRT staging and follow-up. Furthermore, we have shown that RP can be reduced through the use of either V10 in treatment planning or an ACE inhibitor. Finally, esophageal toxicity can be predicted with BED10 to 1.5cc.


This thesis is restricted to Yale network users only. This thesis is permanently embargoed from public release.