/     /     /  

Comparison of Proton and Photon Radiation Treatment Plans for the Adjuvant Treatment of Pancreatic Cancer Including Comprehensive Nodal Radiation

Reviewer: Eric Shinohara MD, MSCI
Abramson Cancer Center of the University of Pennsylvania
Ultima Vez Modificado: 28 de mayo del 2008

Presenter: Mark Ingram
Presenter's Affiliation: University of Pennsylvania
Type of Session: Poster

Background

• Pancreatic cancer is the fourth leading cause of death and most patients present with unresectable disease. The five year survival is poor even in patients who can be resected (5 year survival ~10%) and is even worse in patients who are unresectable (2 year survival ~15%).
• Given these poor outcomes, chemotherapy and radiation can be used to improve local control and overall survival. However, there have been conflicting results regarding the efficacy of radiation in pancreatic cancer. These difference in outcome between studies may be partly due to toxicities related with treatment leading to increased breaks which compromise treatment.
• In order to decrease treatment related side effects, there has been a movement towards reducing treatment volumes and reducing chemotherapy doses of the most active agents in an effort to combine therapy.
• The present study compared various proton beam treatment plans with photon plans delivered using IMRT.

Materials and Methods

• Ten post-operative patients were included in this study.
• Eclipse treatment planning software (Varian) was used to generate treatment plans which encompassed the tumor bed as well as comprehensive nodal irradiation.

Comprehensive nodal irradiation included the porta-hepatic, celiac, superior mesenteric, and peri-pancreatic nodal regions.

• Proton plans used five different field arrangements and the number of beams used ranged from one to four.

Plans were calculated using the Eclipse Proton Convolution Superposition algorithm.

Plans were done using a scatter beam system supplied by Ion Beam Applications (IBA) proton therapy systems.

Compensators and blocks were used to increase conformality with the target volume.

• Photon IMRT plans were generated using 6 and 15 MV photons with five equally spaced coplanar beams.
• The prescription dose was 5,040 cGy in 180 cGy fractions for all plans. Each plan was normalized such that 95% of the clinical target volume (CTV) received 100% of the dose.
• Dose Volume Histograms (DVH) were generated and the volume of the right kidney, left kidney, liver and small bowel that received 20%, 50%, and 80% of the total dose (V20, V50, and V80) were calculated.
• Maximum cord dose was also calculated as well as the maximum and minimum dose delivered to the CTV. The standard deviation of the dose delivered to the CTV was also calculated.

Results

• Kidneys: All five of the proton plans generated lower V20 and V50 values for the left and right kidney compared with the IMRT plans. The greatest sparing was seen in proton plans which did not use posterior beams.
• Liver: All five proton plans generated lower or equal V20 and V50 values compared with the IMRT plans. V80 doses were comparable for the proton and IMRT plans with some IMRT plans having superior V80 compared with proton plans, depending on the proton beam arrangement.
• Small Bowel: V20 to the bowel was comparable for IMRT plans and proton plans which used three or four fields. Proton plans which used two or less fields had lower V20’s. V50 was highest for proton plans which used two or less beams while the three and four beam plans were comparable to the IMRT plans. V80 was found to be highest in the proton plans which used two or less fields.
• Cord: The maximum dose to the cord was less for all proton plans. The four and five field proton plans delivered almost no dose to the cord.
• Target Homogeneity: There was worse homogeneity with the IMRT plans as seen by the larger standard deviation of the CTV doses. There was also a larger difference between the minimum and maximum doses to the CTV with IMRT plans compared with proton plans.

Author's Conclusions

• Proton plans offered superior dose deposition when comprehensive nodal irradiation was delivered along with treatment to the primary tumor
• Due to decreased normal tissue exposure of radiation, radiosensitizers such as gemcitabine may be better tolerated with proton therapy and should be explored.
• Future studies will evaluate the issue of organ motion and targeting of the proton beam and IMRT.

Clinical/Scientific Implications

Pancreatic cancer has an extremely poor prognosis and local recurrences are a major problem. Local control is important not only in the curative setting but also in the palliative setting to prevent pain syndromes due to celiac axis invasion. However, concurrent chemoradiation can be quite toxic and without proper nutritional and fluid support it can be difficult for patients to complete treatment without interruptions. By increasing the conformality of radiation therapy it may be possible to decrease the dose to normal tissues, decreasing toxicity. Additionally, it may allow the dose of radiation used to be increased, improving local control. It may also allow the concurrent use of chemotherapeutics which have a radiosensitizing effect such as Gemzar, which is only rarely used due to combined toxicity. However, with less normal tissue being irradiated it may be possible to deliver radiation concurrently and safely with such agents, which may improve the local effects of radiation as well as decrease distant failures. We may see even less normal tissue toxicity with the use of scanning beam protons and intensity modulated proton therapy (IMPT).

However, clearly greater conformality comes with risks, the most significant being margin misses. There is still breathing motion which must be accounted for in the treatment of pancreatic cancer and with protons, this will become even more important to account for. Clearly clinical trials are needed and as more proton facilities become available these will hopefully move forward.