Biologically adapted radiotherapy and evaluation of non-uniform dose distributions

Radiation therapy plays an integral part in cancer management. Over the last decade, the fraction of patients in the Nordic countries receiving radiation therapy at some stage in their disease has increased by around 50%, and approximately half of the treatments arc given with curative intent. While only 20% of patients with primary tumors receive radiation therapy as the only form of treatment, curative radiation therapy given in combination with other treatment modalities has been shown to be of benefit for the majority of the most common cancers. The future requirements for radiation therapy are expected to increase along with the increase in cancer incidence. The aim of curing the patient is not always achieved, due to distant metastasis and/or lack of locoregional control. Locoregional failure occurs when the delivered tumor dose fails to eradicate the cancer cells, and can result from a radioresistant subpopulation of tumor cells. As the tumor dose is limited by the probability of inducing normal tissue complications, novel treatment strategies are needed to improve locoregional tumor control. Over the recent years, there has been an increasing interest in complex treatment delivery strategies in radiation therapy. Intensity modulated radiation therapy (IMRT) can be used to provide a distribution of radiation dose that conforms closely to irregularly shaped tumors, while sparing the surrounding normal tissues. However, IMRT can also be used to deliver non-uniform dose distributions, based on patient specific biological information, i.e. biologically conformal radiation therapy (BCRT). Functional and molecular imaging can be used to demonstrate both the distribution of and the extent of heterogeneity in parameters that influence tumor radiation sensitivity. Non-invasive images of radiobiological parameters form the basis of biologically conformal radiation therapy and are needed to create individually optimized radiation therapy plans. Recent advances in the imaging of tumor physiological parameters include positron emission tomography (PET) of cellular metabolism and hypoxia, dynamic contrast-enhanced magnetic resonance imaging and computerized tomography (DCEMRI and -CT) of perfusion- and/or permeability-related parameters and blood oxygen level dependent (BOLD) MRI. Conventionally, the desired dose distribution to the tumor has been uniform over the tumor volume. This will optimize the tumor control probability (TCP) in the case of homogeneous tumors, but it will generally not be the optimal dose distribution for tumors with spatial variations in radiation sensitivity. Factors influencing the radiation sensitivity of tumors cells are known to be heterogeneously distributed over the tumor volume. It has been hypothesized that, for a heterogeneous tumor, radiation treatment can be viewed as a selection process, whereby treatment selects for the more resistant subpopulations of tumor cells. Consequently, these cells may come to dominate the response of the tumor to treatment. It follows that, for treatment to be effective, it must successfully eradicate the most resistant fraction of cells in the tumor. However, normal tissue toxicity constrains the radiation dose that can be delivered to the tumor. Hence, in order to maximize TCP for a given mean dose to the tumor, the radiation dose could be redistributed according to the spatial distribution of radiosensitivity in the tumor. This strategy for biological treatment optimization is termed dose redistribution, and could potentially improve locoregional tumor control when treatment response is limited by a radioresistant subpopulation of tumor cells (author)