Biologically guided radiation therapy (BGRT) and the repair-misrepair-fixation (RMF) model

Joo Han Park, Purdue University

Abstract

The dose The dose distribution delivered to a patient on any given treatment day may substantially deviate from the planned dose distribution because of patient setup errors, organ deformation and organ movement. With the advent of new technologies for image-guided radiation therapy (IGRT), periodic (interfraction) dose corrections can be implemented in the clinic to correct for differences among delivered and planned dose distributions. Formulas to determine the size of the biologically-equivalent dose on the n th treatment day needed to correct for systematic as well as random variations in a planned dose distribution on the preceding (n – 1) treatment days are derived using the linear-quadratic (LQ) survival model. The proposed formalism can be used to introduce multiple, periodic inter-fraction corrections or a single post-treatment correction. We show that biologically equivalent corrections for random and systematic deviations from a planned treatment are, under quite general circumstances, always numerically smaller than the difference between the planned and delivered dose to a region of normal or malignant tissue. Also, because of the βGD2 component of cell killing, random errors in the delivered dose do not tend to cancel. Random dose delivery errors have the potential to increase tumor control but may also increase damage to normal tissue structures. Formulas linking LQ cell survival parameters to double strand break (DSB) induction are derived from the Repair-Misrepair-Fixation (RMF) model for mixed radiation fields. Measured data for several cell lines irradiated by radiations of varying quality are used to examine the impact of proximity effects on intra- and inter-track binary misrepair. With the RMF, estimates of α and β for any particle type are determined by a well-defined physical parameter (frequency-mean specific energy), two or three biological parameters, (&thetas; and κ) or (&thetas;, κinter, and κintra), that are independent of radiation quality, and a biological parameter (Σ) that depends on radiation quality. To minimize the number of ad hoc adjustable parameters, we used the published Monte Carlo Damage Simulation (MCDS) to estimate Σ. The effects of radiation quality on α and β were examined by performing a regression analysis of survival data of V79 and CHO cell lines, irradiated with 55 and 250 kVp x-rays and γ-rays from 60Co and human kidney T-1 cells irradiated by x-rays, deuterons and alpha particles with an LET up to 200 keV/μm. In CHO, V79 and T-1 cells irradiated by widely varying types of radiation, the probability per unit time DSB formed by the same track interact in pairwise fashion (κ intra) is 100 to 300 times larger than the probability per unit time DSB formed by different tracks interact in pairwise fashion ( κinter), e.g. the ratio of κ intra and κinter is 235 for CHO. Although the intra-track DSB interaction rate is evidently at least two orders of magnitude larger than the inter-track DSB interaction rate, low LET radiations, such as fast electrons, seldom create more than one DSB per track. For the secondary electrons created in Compton scatter and photoelectric interactions of 60Co γ-rays in or near a cell, less than 0.2% of the 8.3 DSB Gy-1 bp-1 are due to the same electron. Consequently, intra-track DSB interactions are negligible for low LET radiations. However for higher LET radiations, substantial numbers of DSB are produced by a single track and intratrack DSB interactions become the dominate mechanism for pairwise DSB interaction. An attractive feature of the RMF formulas for α and β is that radiosensitivity parameters for intermediate and high LET radiations can be determined with reasonable accuracy from estimates of the cell-specific parameters (&thetas;, κ) derived from an analysis of survival data for low LET radiations (200 and 250 kVp x-rays). Overall, the RMF was found a useful conceptual and mathematical framework to quantify the effects of radiation quality on intrinsic radiation sensitivity for monoenergetic charged particles and mixtures of charged particles of varying quality.

Degree

Ph.D.

Advisors

Poulson, Purdue University.

Subject Area

Medical imaging|Nuclear physics

COinS