The authors report on the local control and toxicity of stereotactic body radiotherapy (SBRT) for patients with unresectable pancreatic adenocarcinoma.
Seventy-seven patients with unresectable adenocarcinoma of the pancreas received 25 gray (Gy) in 1 fraction. Forty-five patients (58%) had locally advanced disease, 11 patients (14%) had medically inoperable disease, 15 patients (19%) had metastatic disease, and 6 patients (8%) had locally recurrent disease. Nine patients (12%) had received prior chemoradiotherapy. Sixteen patients (21%) received between 45 to 54 Gy of fractionated radiotherapy and SBRT. Various gemcitabine-based chemotherapy regimens were received by 74 patients (96%), but 3 patients (4%) did not receive chemotherapy until they had distant failure.

The median follow-up was 6 months (range, 3-31 months) and, among surviving patients, it was 12 months (range, 3-31 months). The overall rates of freedom from local progression (FFLP) at 6 months and 12 months were 91% and 84%, respectively. The 6- and 12-month isolated local recurrence rates were 5% and 5%, respectively. There was no difference in the 12-month FFLP rate based on tumor location (head/uncinate, 91% vs body/tail, 86%; P = .52). The progression-free survival (PFS) rates at 6 months and 12 months were 26% and 9%, respectively. The PFS rate at 6 months was superior for patients who had nonmetastatic disease versus patients who had metastatic disease (28% vs 15%; P = .05). The overall survival (OS) rates at 6 months and 12 months from SBRT were 56% and 21%, respectively. Four patients (5%) experienced grade 2 acute toxicity. Three patients (4%) experienced grade 2 late toxicity, and 7 patients (9%) experienced grade 3 late toxicity. At 6 months and 12 months, the rates of grade 2 late toxicity were 11% and 25%, respectively. SBRT for pancreatic adenocarcinoma was effective for local control with associated risk of toxicity and should be used with rigorous attention to quality assurance. Efforts to reduce complications are warranted. Distant metastases account for the vast majority of disease-related mortality.

Phase I study of stereotactic radiosurgery in patients with locally advanced pancreatic cancer.

Koong AC, Le QT, Ho A, Fong B, Fisher G, Cho C, Ford J, Poen J, Gibbs IC, Mehta VK, Kee S, Trueblood W, Yang G, Bastidas JA. Int J Radiat Oncol Biol Phys. 2004 Mar 15;58(4):1017-21.

Department of Radiation Oncology, Stanford University, Stanford, California 94305-5152, USA. akoong@stanford.edu

To determine the feasibility and toxicity of delivering stereotactic radiosurgery to patients with locally advanced pancreatic cancer. Patients with Eastern Cooperative Oncology Group performance status < or=2 and locally advanced pancreatic cancer were enrolled on this Phase I dose escalation study. Patients received a single fraction of radiosurgery consisting of either 15 Gy, 20 Gy, or 25 Gy to the primary tumor. Acute gastrointestinal toxicity was scored according to the Radiation Therapy Oncology Group criteria. Response to treatment was determined by serial high-resolution computed tomography scanning. RESULTS: Fifteen patients were treated at 3 dose levels (3 patients received 15 Gy, 5 patients received 20 Gy, and 7 patients received 25 Gy). At these doses, no Grade 3 or higher acute gastrointestinal toxicity was observed. This trial was stopped before any dose-limiting toxicity was reached, because the clinical objective of local control was achieved in all 6 evaluable patients treated at 25 Gy. CONCLUSIONS: It is feasible to deliver stereotactic radiosurgery to patients with locally advanced pancreatic cancer. The recommended dose to achieve local control without significant acute gastrointestinal toxicity is 25 Gy.

Normal tissue dosimetry and toxicity in cyberknife radiosurgical treatment of the pancreas
A. Grow, B. Fong, A. Koong  IJROBP 2004:60:S427

Department of Radiation Oncology, Stanford University, Stanford, California

Treatment of unresectable pancreas cancer by conventional chemoradiotherapy has yielded disappointing results. Many deaths are due to local failure, offering the hope that dose escalation may provide better control. This has been difficult to achieve with conventional external beam radiotherapy, due to increased complication probabilities in nearby normal tissues. We have treated patients on a Phase I dose-escalation protocol, using Cyberknife radiosurgery in a single fraction to deliver highly conformal doses to unresectable pancreas tumors with minimal dose to surrounding normal tissues. We studied the relationship between dose to various tissues and incidence of GI toxicity.  Eighteen patients were treated at three dose levels: 15, 20 or 25 Gy prescribed to the tumor in a single fraction. All patients had unresectable adenocarcinoma of the pancreas with maximum tumor size 7.5 cm. Study entry criteria also included ECOG PS of 2 or better, and no chemotherapy within the prior two months. One of the 18 patients had had prior external beam radiotherapy. The average time for which patients were followed and monitored for toxicity was 9 months (time from treatment to death or to this analysis). Three patients were treated at the 15-Gy dose; of these, one had Grade II GI toxicity (by RTOG criteria). Zero of 6 patients treated at 20 Gy and four of 9 patients treated at 25 Gy had either Grade I or Grade II toxicity, for a total of 5 patients in the toxicity group and 13 in the no-toxicity group. To analyze the relationship between organ dose and toxicity, dose-volume histograms were generated for each patient. Contours were drawn on the thin-cut planning CT scan for the tumor as well as for the remaining normal pancreas, duodenum, other bowel, stomach, liver, kidneys, spleen, aorta, inferior vena cava and spinal cord. A body imaging radiologist guided us in localizing the duodenum and other difficult structures. Dose-volume histograms were generated, and doses to duodenum, stomach and other bowel (organs relevant for GI toxicity) were studied using differential area under the curve (integrated area under the dose-volume curve for the portion of the organ receiving a dose greater than 100 cGy) as a parameter.

No patients experienced GI toxicity exceeding Grade II. To evaluate whether the likelihood of toxicity was related to prescribed tumor dose within the studied range (15–25 Gy), we calculated the proportion of patients receiving the high dose of 25 Gy in the toxicity group (4/5 = 0.8) and in the no-toxicity group (5/13 = 0.385). When these proportions were compared using the Z statistic, the result was nonsignificant (p > 0.05). To evaluate effects of dose to individual organs, the average differential area under the curve was calculated for each relevant organ in the toxicity and no-toxicity groups. When the results were compared between groups using the t test, we found that the difference was significant (p < 0.05) for the duodenum, but not for the stomach or other bowel. The lack of Grade III or higher toxicities indicates that our maximum dose (25 Gy) is a safe single-fraction dose. We found that patients who had Grade I-II toxicity were not significantly more likely to have been treated at the maximum dose than those who had no toxicity, indicating that the likelihood of toxicity may be related more to the proximity of the tumor to certain normal tissues than to the prescribed tumor dose. In our analysis of dose-volume data for organs relating to GI toxicity, we found that the likelihood of toxicity was significantly related to the duodenal dose, although not to the stomach or remaining bowel doses. From this we conclude that duodenal dose may be the most important determinant for Grade I-II toxicities, and that it may be possible to establish a simple DVH-related parameter to guide radiosurgical treatment planning in avoidance of acute and subacute GI toxicity. Analysis is ongoing for a Phase II study involving a combination of external beam RT followed by radiosurgery, and we hope to be able to draw further conclusions concerning both more severe and late toxicities