Radiation and Cancer Physics

**SS 27 - Physics 7 - Special Session: Outcome Analysis and Modeling**

# 198 - Estimation of the Alpha-beta Ratio for Chemoradiation of Locally Advanced Pancreatic Cancer

Tuesday, October 23

4:05 PM - 4:15 PM

Location: Room 214 C/D

** Estimation of the Alpha-beta Ratio for Chemoradiation of Locally Advanced Pancreatic Cancer**

__P. W. Prior Jr__^{1}, X. Chen^{1}, W. A. Hall^{2}, B. A. Erickson^{1}, and A. Li^{1}; ^{1}Medical College of Wisconsin, Milwaukee, WI, ^{2}Medical College of Wisconsin and Clement J Zablocki VA Medical Center, Milwaukee, WI

**Purpose/Objective(s): **Different fractionations including stereotactic body radiation therapy (SBRT) and conventionally fractionated radiotherapy (CFRT) are used in chemo-radiotherapy (CRT) for locally advanced pancreatic cancer (LAPC). A plausible radiological parameters (*e.g.*, α/β) are desirable for comparing and desiring different RT fractionations. The purpose of our work is to estimate a set of these parameters based on biophysical modeling of tumor response rate (RR) and tumor control probability (TCP) using published clinical data. .

**Materials/Methods: **A literature search for CRT using SBRT and CFRT treatments was performed where CT-based RR (complete and partial) data were reported. A modified linear quadratic models (mLQ) were used to model RR that considers a Gaussian distribution of the critical volume required to achieve tumor response according to RECIST criteria, biologically effective dose (BED), and tumor repopulation, The RR model includes 5 radiobiological parameters: *A* – the ratio of the total number of tumor cells to the standard deviation (STD) in tumor cells, *B* – the ratio of the critical tumor volume for 50% of nonresponding cells to the STD in tumors cells, *α*, *α/β*, *fc* - to account for chemotherapy effect. Published clinical data were used to fit these models and the derived parameter set was used to calculate BED for different fractionation schemes.

**Results: **A literature search yielded 28 studies (944 patients) with reported CRT based tumor response data. A wide range of fractionations was used and ranged from 1.8 Gy per fraction (Gy/fx) (*n*=21), 3-9 Gy/fx (*n*=5), and 10-25 Gy/fx (*n*=2). A spread of TCP as a function of BED was observed. The mLQ-TCP model fitting parameters were found to be *A* = 4.01±0.92, *B* = 0.29±0.78, *α*=0.015±0.003 Gy^{-1}, *α/β* =9.5±5.6 Gy, and *fc* = 1.316±0.060 for Gemcitabine based CRT. The previously published value of fc=1.188 was used for 5FU CRT. The goodness of the fit was χ^{2}/dof = 1.52. Based on the the obtained model parameters, fractionations with 5.17, 10 and 25 Gy/fx results in a comparable BED with 57.25 Gy in 25 fractions.

D (Gy) | d (Gy/fraction) | T (days) | BED9.5 (Gy) |

50.0 | 10.00 | 7.0 | 149.3 |

50.4 | 1.80 | 38.0 | 41.1 |

57.3 | 2.29 | 35 | 60.7 |

77.5 | 5.17 | 21.0 | 151.9 |

25.0 | 25.00 | 1.0 | 145.7 |

**Conclusion: **The tumor control rate is clearly related to the radiation dose and fractionation in both SBRT and CFRT for LAPC based on the clinical reports reported, which can be described by the mLQ TCP model. A plausible set of radiosensitivity parameters (*e.g.*, α/β = 9.5) along with their ranges were obtained. They can be useful in desiring physical and biological conformal dose fractionation schemes to take advantages of MRI-guided dose painting for pancreatic cancer.

__Author Disclosure:__ **P.W. Prior**: None. **X. Chen**: None. **W.A. Hall**: None. **B.A. Erickson**: Employee; ProHealth Care, Waukesha, WI. Steering committee; American Brachytherapy Society. **A. Li**: None.