Ten years of progress in radiation oncology

Computed tomography (CT) scans acquired in the radiotherapy treatment position before the start of radiotherapy remain the basic imaging modality for contouring tumor target volumes and healthy tissues ("organs at risk") as well as for dose calculation in radiotherapy planning. As dose-response relationships have been demonstrated for several tumor types ("higher dose to the tumor = better chance of cure"), for example, in radical radiotherapy of prostate cancer or non-small-cell lung cancer, efforts to increase the radiotherapy dose in limited tumor volumes with small margins were undertaken. However, the inability of CT to provide functional information, for example, on tumor vitality, proliferation, oxygenation or perfusion and the problem of day-to-day organ motion have necessitated additional information to advance radiotherapy planning.

Functional imaging modalities, such as magnetic resonance imaging spectroscopy (MRS), and, in particular, positron-emission tomography (PET) have opened new possibilities to obtain metabolic information and identify the most radioresistant subvolumes within a tumor [2]. MRS-defined dominant tumor lesions, for example, in the prostate, can be specifically addressed by an escalated radiotherapy dose [3].

Extremely precise delivery of high radiation doses to small volumes was already technically possible in the 1990s and favorable results were obtained in benign and malignant brain tumors with a few fractions ("hypofractionated") or single-fraction stereotactic radiotherapy ("radiosurgery") [4]. The main indications for this technique are brain metastases, recurrent (previously irradiated) malignant gliomas, vestibular schwannomas and meningiomas. The brain is ideal for this procedure, as tumor or organ motion is practically non-existent.

The problem of motion of tumor-bearing organs as well as adjacent healthy organs, most prominently exemplified by day-to-day motion of the prostate due to varying filling states of the rectum and of lung tumor movement within the breathing cycle, has been addressed by the implementation of image-guided radiotherapy (IGRT). Whereas only boney structures could be visualized in the past on the treatment couch of the linear accelerator at the time of each radiotherapy fraction, the integration of computed tomography into linear accelerator technology ("cone-beam CT") as well as the option to introduce radio-opaque fiducial markers into tumors or tumor-bearing organs, such as the prostate (Figure 1), made possible the correction of the patient position based on this information at each treatment session, thereby drastically reducing the margins around the tumor/organ required to compensate for motion.

Such advanced imaging on the treatment table was a prerequisite for the clinical introduction of advanced algorithms of dose calculation and delivery. Intensity-modulated radiotherapy (IMRT) enabled radiation physicists to create treatment plans with highly individualized dose distributions and a sharp dose gradient at the interface of tumor volume and healthy organ, even if the latter is virtually enclosed by the former [5]. Typical examples include the sparing of the highly radiosensitive parotid glands in radiotherapy of head and neck cancer and the protection of rectal mucosa adjacent to prostate and seminal vesicles (Figure 2). Sophisticated target volumes based on functional imaging data, IGRT and IMRT have been integrated in novel radiotherapy concepts [6]. Tomotherapy, an advanced type of IMRT, integrates imaging of the patient and delivery of radiotherapy in a sectional manner [7].

Proton radiotherapy, due to advantageous physical properties, has the potential to further improve clinical results so far achievable with modern linear accelerator photon radiotherapy. Like recent improvements in photon delivery, increased (biologically effective) doses in the tumor volume and/or reduced radiation dose in healthy organs - as achievable with protons according to theoretical planning studies - may further improve the therapeutic ratio of radiotherapy. However, clinical trial data is needed to fully assess the potential of proton radiotherapy [8].

In the past, based on the results of large randomized trials and meta-analyses, specific recommendations for the delivery of radiotherapy were made for tumor entities and disease stages. Even today, such statements in national and international guidelines for cancer treatment define the standards of care. However, the assessment of tumor material in individual patients has been proposed as a predominant source of information on which to base treatment decisions. Specific combinations of biomarkers detectable by immunohistochemistry (tissue microarrays) and specific gene signatures detectable in gene microarray studies have been used predominantly to predict the benefit from postoperative chemotherapy. While a focus of this field has been to identify subgroups of breast cancer patients benefiting from particular types of systemic therapy, response to radiotherapy has equally been addressed by microarray studies, for example, in diseases treated with radical radiotherapy such as cervix cancer [9].

Experimental studies of radiosensitivity of tumor cells in in-vitro and in-vivo models have identified important mechanisms of radioresistance. Some of these findings could already be translated into clinically useful protocols of radiotherapy in combination with molecular targeting agents. The most prominent example is targeting of the epithelial growth factor receptor (EGFR) in combination with radiotherapy. Initially, the association of EGFR overexpression with prognosis was assessed in several tumor types [10]. In a randomized trial in head and neck cancer, EGFR targeting improved the outcome compared to radiotherapy alone, leading to further trials of treatment intensification with more complex drug combinations as well as to new translational research initiatives [11].

Low tumor oxygenation is a frequently observed cause of poor response to radiotherapy, for example, in head and neck or cervix cancer. Normalizing tumor oxygenation and specifically targeting or radiosensitizing hypoxic tumor cells have been alternative strategies to improve the tumor control rates in hypoxic tumors. Recently, hypoxia-related molecules have been assessed as targets in combination with radiotherapy, showing some potential for radiosensitization of tumor cells [12].