Modelling beam transport and biological effectiveness to develop treatment planning for ion beam radiotherapy

TL;DR

This paper develops a treatment planning algorithm for carbon ion radiotherapy that accurately models biological effectiveness and beam transport to optimize tumor cell inactivation.

Contribution

It introduces a novel treatment planning system kernel combining a validated radiobiological model with a physical beam transport model for improved ion therapy planning.

Findings

Validated Katz's cellular survival model for ion fields.

Created an optimization tool for uniform cell inactivation.

Integrated physical and biological models for treatment planning.

Abstract

Radiation therapy with carbon ions is a novel technique of cancer radiotherapy, applicable in particular to treating radioresistant tumours at difficult localisations. Therapy planning, where the medical physicist, following the medical prescription, finds the optimum distribution of cancer cells to be inactivated by their irradiation over the tumour volume, is a basic procedure of cancer radiotherapy. The main difficulty encountered in therapy planning for ion radiotherapy is to correctly account for the enhanced radiobiological effectiveness of ions in the Spread Out Bragg Peak (SOBP) region over the tumour volume. In this case, unlike in conventional radiotherapy with photon beams, achieving a uniform dose distribution over the tumour volume does not imply achieving uniform cancer cell inactivation. In this thesis, an algorithm of the basic element (kernel) of a treatment planning system (TPS) for carbon ion therapy is developed. The algorithm consists of a radiobiological part which suitably corrects for the enhanced biological effect of ion irradiation of cancer cells, and of a physical beam transport model. In the radiobiological component, Katz's track structure model of cellular survival is applied, after validating its physical assumptions and improving some aspects of this model. The Katz model offers fast and accurate predictions of cell survival in mixed fields of the primary carbon ions and of their secondary fragments. The physical beam model was based on available tabularized data, prepared earlier by Monte Carlo simulations. Both components of the developed TPS kernel are combined within an optimization tool, allowing the entrance energy-fluence spectra of the carbon ion beam to be selected in order to achieve a pre-assumed uniform (flat) depth-survival profile over the SOBP region, assuring uniform cancer cell inactivation over the tumour depth.