Radiation-induced second cancers in prostate cancer patients

Radiotherapy is a commonly used, potentially curative treatment modality in patients with localised and locally advanced prostate cancer. The development of a radiation-induced second primary cancer is a serious long-term side effect of radiotherapy treatment. Prostate cancer is the most common cancer in males in the UK, with a lifetime risk of one in eight. Since the advent of prostate specific antigen (PSA) testing, patients are diagnosed at an earlier stage in their disease and patients are surviving for longer following the diagnosis. The risk of radiation-induced second cancer is therefore particularly relevant for this group of patients. Traditionally, a radiation-induced second cancer is considered as one which: 1) develops at least five years following irradiation 2) is of a different histological type to the original primary 3) was not present at the time of diagnosis of the original primary, and 4) develops within the field of radiation. For prostate cancer, regions considered within the radiation field include the rectum, bladder, anus and the soft tissues and bones of the pelvis. Two important caveats, however, must be borne in mind when considering the diagnosis of radiation-induced cancer. Firstly, patients may develop second cancers as a result of genetic and environmental factors (eg smoking or dietary factors) rather than due to radiotherapy. As such it should not automatically be assumed that a previously irradiated patient who develops a second cancer has done so because of the radiotherapy. Secondly, although the traditional diagnosis of a radiationinduced second cancer is one which develops in an ‘in-field’ region, patients are exposed to radiation doses far beyond the field edge (traditionally considered as the 50% isodose), and these lower doses could also result in radiation-induced second cancers. The vast majority of the clinical evidence regarding second cancer risk from radiotherapy is retrospective. Studies comparing second cancer risk in irradiated prostate cancer patients and prostate cancer patients managed with other techniques (eg surgically treated patients) are considered here. Most studies examine the risk of second rectal and/or bladder cancers as these are the most commonly identified second cancers observed to occur within the radiation field in prostate cancer patients. Studies require very large numbers of patients in order to have sufficient power to detect real differences between patient groups. Such large patient numbers are often available from registry data (eg SEER – Surveillance Epidemiology and End Results databases) which generally contain tens of thousands of patients. One of the problems with registry data, however, is that the data is often less complete than institutional databases. For example, registry data often does not contain information about smoking status, and so important confounders may be missed. Registry data is often less complete in older patients. Data from single institutions may often contain more details about potential confounding factors, but the far smaller patient numbers mean that the power of these studies is more limited in detecting true differences between patient groups.