31cd. ()*u'oL ,_~ Tum*." l'/tamptacolh{'r. Vol. S. No, 3. pp. 14 I- 145. t t)g I Printed in (}lc~t l~ritain
CANCER
RISKS AFTER
{1736-tll I~.~/t)l $3.(R} + .t/O Pergamon Prc-s ptc
MEDICAL
RADIATION
PER F, L. }IALL Department of General ()neology, Radiumhemnlet, Karolinska 1lospitat, S- t{14t}l Stockholm, Sweden ( l & c c i r e d 5 Mflv l~)91 : ~lccepl~'d 7 M a y 1991)
Radiation-induced t u m o r s c a n n o t be distinguished from t u m o r s in general hy means o t h e r t h a n a statistical excess. Epidemiological stndies are the only nleans by which answers can be given as r e g a r d s the carcinogenic effect of ionizing radiation. Age at exposure is p e r h a p s the most i m p o r t a n t host factor inllttencing c a n c e r risk a n d it is generally believed that c a n c e r risk decreases with increasing age at exposure. For most cancers the t e m p o r a l p a t t e r n foUows the n a t u r a l incidence~ i.e. the cancers do not occur before ages n o r m a l l y associated with increased incidence, T h e induction period for solid t u m o r s is at least it) years while the correspoudint2 I]gure l~r leukemia is 2 years. The breast, thyroid, [ling a n d bone m a r r o w seem to be the most radiosensitive tissues, while the risk of chronic lymphatic leukemia and possibly H o d g k i n ' s disease a n d prostatic c a n c e r does not seem to increase after exposure to ionizing radiation. Key n'ortLs: C a n c e r risks, Ionizing radiation, Induction period, Age at exposure.
tuberculosis were normal in all other health respects related to cancer, for instancc socio-economic status. Cancer patients also often receive other treatments that may affect the cancer risk. Most of them are also elderly and have fewer remaining years of life under risk. A third problem is the increased medical surveillance of cancer patients which mcrcascs the chance of detecting cancers not yct showing symptoms, On the other hand, populations exposed to medical radiation usually have adequate data on dosimetry and follow-up. Two models are discussed for the mechanisms of carcinogenesis. These models have different practical implications when risks to other exposed populations are computed or when extra-polating risks to other dose levels, l The multiplicative model means that the relative risk is constant regardless of age, i,e, the harmful effect of ionizing radiation is concentrated in those with the highest underlying cancer risks such as the elderly. In the additive or absolute risk model the cancer risk is independent of age, None of these risk models reflects the exact truth. The type of tissue exposed, age at exposure, and sex are some of the factors that modify the interpretation of the carcinogenic effect, ~
INTRODUCTION The radiation dose received from natural sources is by far the largest contributor to the collective dose received by the world population. Medical exposures to ionizing radiation constitute the largest man-made contributi0n, and approach the dose received from natural sources in some industrialized countries. ~ Populations exposed to ionizing radiation in medical situations are some of the major sources from which data on the carcinogenic effects of ionizing radiation can be drawn. ~'2 There are several problems related to the studies of carcinogenesis in patients. The dose delivered is often higher than the dose received by most individuals during a life-time. T h i s introduces the problem of interpolating high radiation doses to lower dose levels. Much information has, during the past years, been added to our knowledge on highdose exposure; d o s e - r e s p o n s e patterns have been made more accurate by combining different studies, 3 dose estimates have been revised, ~'5 older studies have been reanalyzed thus adding person years of observation. Despite these gains, the shape of the dose-response curve at low levels of radiation and the effect of low doses are still uncertain. 6 The exposed population is highly selected insofar as many of the individuals were exposed because of an underlying disease and have an age distribution different from the population as a whole, It is not obvious that patients suffering from tinea capitis or
MATERIALS AND M E T H O D S The cohorts listed in Table 1 are some of the 141
142
Per F. L~ Hall Table t. Populations discussed
Diagnostic procedures: Lung fluoroscopy Thyroid examination Radiotherapy for benign diseases: Ankylosing spondytitis Thymic enlargement Tinea capitis Postpartum mastitis Hyperthyroidism Radiotherapy for malignant diseases: Cervical cancer Thyroid cancer
studies of patients exposed to ionizing radiation. They are briefly discussed below.
Diagnostic procedures Millions of individuals have been exposed to low levels of ionizing radiation through diagnostic procedures. With the improvement of diagnostic techniques and equipment, doses are being reduced. The availability of ultrasound and magnetic resonance imaging will also reduce the use of ionizing radiation. ~ However, the increasing use of computer tomography and the various uses of radioisotopes for diagnostic purposes may have increased the total dose received, t
Pneumothorax fluoroscopy. Multiple chest fluoroscopies to monitor pneumothorax treatment of tuberculosis were used in 6,285 patients between 1925 and 1954 in Massachusetts. 7's Significantly elevated risks of breast and esophageal cancer were seen. For esophageal cancer the risk decreased with time since exposure, The overall cancer mortality was not increased. No significantly excessive death from leukemia was seen with a mean dose to the active bone marrow of 0.09 Gy. This information is of particular importance since the dose was delivered over a period of several years and at low doses, and the situation thus resembles the exposure in occupational settings more than in individuals exposed to high, single doses. Thyroid examination. Cancer risks in more than 35,000 patients receiving diagnostic doses of radioactive iodine for a suspected thyroid disease were studied, and statistically significant increases were seen for endocrine tumors other than thyroid cancer, tymphomas, leukemia, and nervous system tumors. '~ It was concIuded that the increases noted were due
to selection factors since the combined risk was highest 5-9 years after exposure, and did not differ from unity after l0 years of follow-up. However, the significant risk of leukemia, other than CLL, warrants special attention since it is considered a prominent radiogenic malignancy. The close to the bone marrow in this study was
CANCER
RISKS AFTER
{1736-tll I~.~/t)l $3.(R} + .t/O Pergamon Prc-s ptc
MEDICAL
RADIATION
PER F, L. }IALL Department of General ()neology, Radiumhemnlet, Karolinska 1lospitat, S- t{14t}l Stockholm, Sweden ( l & c c i r e d 5 Mflv l~)91 : ~lccepl~'d 7 M a y 1991)
Radiation-induced t u m o r s c a n n o t be distinguished from t u m o r s in general hy means o t h e r t h a n a statistical excess. Epidemiological stndies are the only nleans by which answers can be given as r e g a r d s the carcinogenic effect of ionizing radiation. Age at exposure is p e r h a p s the most i m p o r t a n t host factor inllttencing c a n c e r risk a n d it is generally believed that c a n c e r risk decreases with increasing age at exposure. For most cancers the t e m p o r a l p a t t e r n foUows the n a t u r a l incidence~ i.e. the cancers do not occur before ages n o r m a l l y associated with increased incidence, T h e induction period for solid t u m o r s is at least it) years while the correspoudint2 I]gure l~r leukemia is 2 years. The breast, thyroid, [ling a n d bone m a r r o w seem to be the most radiosensitive tissues, while the risk of chronic lymphatic leukemia and possibly H o d g k i n ' s disease a n d prostatic c a n c e r does not seem to increase after exposure to ionizing radiation. Key n'ortLs: C a n c e r risks, Ionizing radiation, Induction period, Age at exposure.
tuberculosis were normal in all other health respects related to cancer, for instancc socio-economic status. Cancer patients also often receive other treatments that may affect the cancer risk. Most of them are also elderly and have fewer remaining years of life under risk. A third problem is the increased medical surveillance of cancer patients which mcrcascs the chance of detecting cancers not yct showing symptoms, On the other hand, populations exposed to medical radiation usually have adequate data on dosimetry and follow-up. Two models are discussed for the mechanisms of carcinogenesis. These models have different practical implications when risks to other exposed populations are computed or when extra-polating risks to other dose levels, l The multiplicative model means that the relative risk is constant regardless of age, i,e, the harmful effect of ionizing radiation is concentrated in those with the highest underlying cancer risks such as the elderly. In the additive or absolute risk model the cancer risk is independent of age, None of these risk models reflects the exact truth. The type of tissue exposed, age at exposure, and sex are some of the factors that modify the interpretation of the carcinogenic effect, ~
INTRODUCTION The radiation dose received from natural sources is by far the largest contributor to the collective dose received by the world population. Medical exposures to ionizing radiation constitute the largest man-made contributi0n, and approach the dose received from natural sources in some industrialized countries. ~ Populations exposed to ionizing radiation in medical situations are some of the major sources from which data on the carcinogenic effects of ionizing radiation can be drawn. ~'2 There are several problems related to the studies of carcinogenesis in patients. The dose delivered is often higher than the dose received by most individuals during a life-time. T h i s introduces the problem of interpolating high radiation doses to lower dose levels. Much information has, during the past years, been added to our knowledge on highdose exposure; d o s e - r e s p o n s e patterns have been made more accurate by combining different studies, 3 dose estimates have been revised, ~'5 older studies have been reanalyzed thus adding person years of observation. Despite these gains, the shape of the dose-response curve at low levels of radiation and the effect of low doses are still uncertain. 6 The exposed population is highly selected insofar as many of the individuals were exposed because of an underlying disease and have an age distribution different from the population as a whole, It is not obvious that patients suffering from tinea capitis or
MATERIALS AND M E T H O D S The cohorts listed in Table 1 are some of the 141
142
Per F. L~ Hall Table t. Populations discussed
Diagnostic procedures: Lung fluoroscopy Thyroid examination Radiotherapy for benign diseases: Ankylosing spondytitis Thymic enlargement Tinea capitis Postpartum mastitis Hyperthyroidism Radiotherapy for malignant diseases: Cervical cancer Thyroid cancer
studies of patients exposed to ionizing radiation. They are briefly discussed below.
Diagnostic procedures Millions of individuals have been exposed to low levels of ionizing radiation through diagnostic procedures. With the improvement of diagnostic techniques and equipment, doses are being reduced. The availability of ultrasound and magnetic resonance imaging will also reduce the use of ionizing radiation. ~ However, the increasing use of computer tomography and the various uses of radioisotopes for diagnostic purposes may have increased the total dose received, t
Pneumothorax fluoroscopy. Multiple chest fluoroscopies to monitor pneumothorax treatment of tuberculosis were used in 6,285 patients between 1925 and 1954 in Massachusetts. 7's Significantly elevated risks of breast and esophageal cancer were seen. For esophageal cancer the risk decreased with time since exposure, The overall cancer mortality was not increased. No significantly excessive death from leukemia was seen with a mean dose to the active bone marrow of 0.09 Gy. This information is of particular importance since the dose was delivered over a period of several years and at low doses, and the situation thus resembles the exposure in occupational settings more than in individuals exposed to high, single doses. Thyroid examination. Cancer risks in more than 35,000 patients receiving diagnostic doses of radioactive iodine for a suspected thyroid disease were studied, and statistically significant increases were seen for endocrine tumors other than thyroid cancer, tymphomas, leukemia, and nervous system tumors. '~ It was concIuded that the increases noted were due
to selection factors since the combined risk was highest 5-9 years after exposure, and did not differ from unity after l0 years of follow-up. However, the significant risk of leukemia, other than CLL, warrants special attention since it is considered a prominent radiogenic malignancy. The close to the bone marrow in this study was
