Morbidity and mortality following pancreas SBRT

Committee/Faculty

Presenting Author(s)

SE

Susannah G. Ellsworth, n/a

M.D.
University of Pittsburgh Hillman Cancer Center
Pittsburgh, Pennsylvania, United States

Disclosure: I do not have any relevant financial / non-financial relationships with any proprietary interests.

Co-Author(s)

AD

Annissa DeSilva

B.S.
University of Pittsburgh Hillman Cancer Center
Pittsburgh, Pennsylvania, United States

Disclosure: Disclosure information not submitted.
SB

Steven Burton

M.D.
University of Pittsburgh Hillman Cancer Center
Pittsburgh, Pennsylvania, United States

Disclosure: Disclosure information not submitted.
AO

Adam Olson

M.D.
University of Pittsburgh Hillman Cancer Center
Pittsburgh, Pennsylvania, United States

Disclosure: Disclosure information not submitted.
KL

Kenneth Lee

M.D.
University of Pittsburgh Hillman Cancer Center
Pittsburgh, Pennsylvania, United States

Disclosure: Disclosure information not submitted.
NB

Nathan Bahary

M.D., Ph.D
University of Pittsburgh Hillman Cancer Center
Pittsburgh, Pennsylvania, United States

Disclosure: Disclosure information not submitted.
MA

Mohamed Abdelhakeim

M.D.
University of Pittsburgh Hillman Cancer Center
Pittsburgh, Pennsylvania, United States

Disclosure: Disclosure information not submitted.
AP

Alessandro Paniccia

M.D.
University of Pittsburgh Hillman Cancer Center
Pittsburgh, Pennsylvania, United States

Disclosure: Disclosure information not submitted.
AZ

Amer Zureikat

M.D., F.A.C.S.
University of Pittsburgh Hillman Cancer Center
PIttsburgh, Pennsylvania, United States

Disclosure: Disclosure information not submitted.

Purpose: Stereotactic body radiotherapy (SBRT) can optimize local control in both resected and inoperable pancreatic cancer (PC). Short- and medium-term morbidity and mortality rates following SBRT for PC have not been previously reported. We analyzed acute morbidity and mortality rates and incidence of late vascular toxicity (GI bleeding and visceral pseudoaneurysm) following SBRT for PC.

Methodology: Outcomes were abstracted from a single-institution registry-based series of 464 patients. Hospital and clinic records were reviewed to identify death date/cause of death, any hospital admission within 90 days of the last RT fraction, and instances of upper GI bleeding at any time following SBRT. Survival was calculated from the last day of radiation.

Results: Median age was 69 years (range 32-91); 49.8% of patients were women, and the majority (93%) had adenocarcinoma. SBRT was performed perioperatively in 184 patients (59/12.4% neoadjuvant and 125/26.4% adjuvant); the remainder had definitive (202/42.7%) or salvage (44/9.3%) treatment. Most patients received 3 fractions (n = 301, 63.5%); 91 (19.2%) received single-fraction SBRT, and 72 patients (15.2%) received 5 fractions.
Nine patients (1.9%) died within 30 days of SBRT. Cause of death was unknown in 3 patients, with 3 dying of progressive disease, 2 of infection, and 1 of pulmonary embolism. Twenty-six patients (5.1%) died between 31 and 90 days after SBRT. Cause of death was unknown in 9; 12 died of disease progression, 4 of infection, and 1 of GI bleed. SBRT type (perioperative vs. definitive), patient age, and prior radiation to the pancreas were not correlated with early mortality risk. Thirty-eight patients (8.2%) were admitted within 30 days of SBRT, and 63 (13.6%) were admitted between 31 and 90 days. The most common admission diagnosis during both the first 30 and 90 days after RT was infection, followed by GI bleeding and bowel obstruction.
For the whole cohort, crude GI bleed risk following SBRT was 13.3% (62/464) and was significantly correlated with fractionation scheme (Figure 1). After excluding patients who were lost to follow-up, crude rates of GI bleeding were 22.6% (12/53), 18.8% (47/250), and 4.5% (3/66) in patients treated with 1, 3, and 5 fractions, respectively.; compared with patients undergoing 1- and 3-fraction regimens, respectively, HR for GI bleeding in patients treated with 5 fractions was 0.172 (95% CI 0.046-0.645) and 0.205 (95% CI 0.062-0.681). Six patients (1.3%) developed visceral pseudoaneurysm.

Conclusions: SBRT for PC is associated with relatively low rates of acute morbidity and mortality, although infection and early disease progression are important causes of hospitalization and death in this medically fragile population. GI bleeding risk is highly sensitive to RT fractionation regimen. Dosimetric analysis is ongoing to identify additional dose-volume parameters that are correlated with GI bleeding risk.

References: