Advances in radiotherapy treatment.

It has always been possible to kill all cancers with radiotherapy--the critical point is to avoid killing the patient at the same time.

There are two ways of reducing the difference between cancer damage and normal tissue damage--the first is anatomical, with better target definition and radiation delivery, and the second is using the biological differences with fractionating therapy. I will start with the latter.

Radiobiological advantages to Fractionation 

Radiation damage manifests in two ways:

* acute damage to fast-dividing tissues such as mucosa--this is unpleasant but not usually life threatening

* late damage--this is due to constriction of small blood vessels that results in organ failure 1--10 years after radiation and may cause death, e.g. heart failure.

Dividing the radiation dose into smaller fragments specifically reduces the late effects--the reason for the standard fractionation of 60+Gy in 30+ fractions over 6 weeks. Several approaches have been used to try to exploit these advantages, from twice-daily small fractions to accelerated courses of treatment, but very few of these have shown major clinical benefits and most are impractical in busy radiotherapy departments.

Changing radiation sensitivity

* Increasing the sensitivity of the tumour to radiation is obviously attractive and there are several chemotherapy agents that have such an effect at relatively low doses, e.g. cisplatin and 5FU. Concurrent chemoradiation is also becoming important as there seems to be a synergistic effect between the chemo and the radiotherapy but care must be taken in these settings as the toxicity of the treatment also increases.

* Reducing normal tissue sensitivity to radiation--the only agent currently licensed for this use is amifostine and the results, unfortunately, have been disappointing.

Better anatomical definition of the target and organs at risk (OAR)

Over the last few years there have been enormous advances in imaging techniques and clarity. This enables not only better definition of the target but also identification of OARs. Combining different imaging modalities may also offer advantages--showing extension of tumour in one modality that may not easily be seen in another. This is particularly useful with biological based imaging, e.g. positron emission tomography (PET) or magnetic resonance spectroscopy that may be very sensitive but not have a clear anatomical specificity. If these images can be fused (using computer software) to an image modality that has high anatomical specificity but relatively poor biological specificity we are able to adapt our treatment portals to cover the area of disease far more accurately without unduly increasing complications by using classic 'extended fields' that encompassed all 'areas at risk'.

All current radiotherapy planning systems are based on 3-dimensional tomographic imaging (usually thin-slice CT scans) and most are able to also incorporate and fuse other image series (e.g. MRI, or even preoperative images if these are relevant).

Better systems of radiation delivery

* The initial advances in 'non-coplanar' or 'highly conformal' radiotherapy involved the ability to closely conform the radiation field to the shape of the tumour shielding normal tissue. Initially this was performed with individually made lead blocks but later multileaf collimators were developed. By altering the position of the gantry (radiation source) and the couch it also became possible to treat the tumour from a variety of different angles, again improving the ability to reduce the dose to normal tissue. After these advances there were numerous studies, initially in prostate cancer, that showed that the dose to the prostate could be escalated significantly, improving local control without increasing toxicity. These clinical results have been confirmed in several other tumours, particularly non-small-cell lung cancer and head and neck cancer.

* Quality control and electronic portal imaging (EPID)--improvements in flat-screen technology have allowed the introduction of electronic check films of the patient on the treatment bed, in the treatment position, and allows these to be repeated as often as needed.

* Radiosurgery--very tightly focused radiation given, often in a single fraction, to well-defined intracranial targets, e.g. acoustic neuromas, arteriovenous malformations (AVMs) with accuracy of 0.1 mm using stereotactic techniques.

* Intensity modulated radiation therapy (IMRT). This technique has become extremely popular over the last few years and involves changing not only the shape of the beam but also the intensity of the dose given across the beam. It is highly dependent on complicated computer planning (inverse treatment planning)-where the planner can determine minimum or maximum clinically acceptable doses to specified organs or points. The clinical advantage (or disadvantage) of this approach is yet to be proven and the risks of induction of second malignancy may be higher.

* Image-guided radiotherapy (IGRT) and adaptive radiotherapy (ART) are methods to actually visualise the position of the target within the patient and to adapt the treatment plan using this information, possibly even on a daily basis. This may be achieved by implanted fiducials and orthogonal X-rays, ultrasound localisation or tomography.

* Tomotherapy involves a rotating radiation source as in a CT scan--this will allow real-time measurement of the dose actually delivered. Research units are treating patients.

* Charged particle therapy. The physics of charged particles is such that there is always a lower entrance dose and, after a bragg peak at a depth specific to the particle, there is no exit dose. Thus the dose distribution with these modalities will always be better than photons. There are currently several small hospital-based proton facilities and several units using heavy ions (such as C) but these are very expensive to set up and run.

Jenny WILSON, MB ChB, FFRad (Onc) Professor, Department of Radiation Oncology, Pretoria Academic Hospital