Leukemia effected blood cells can be viewed using a microscope.Leukemia is a cancer of the early blood-forming cells. Usually, the leukemia is a cancer of the white blood cells, but leukemia can involve other blood cell types as well. Leukemia starts in the bone marrow and then spreads to the blood. From there it can go to the lymph nodes, spleen, liver, central nervous system (the brain and spinal cord), testes (testicles), or other organs. Leukemia is among the most likely forms of malignancy resulting from overexposure to total body radiation. Chronic lymphocytic leukemia does not appear to be related to radiation exposure.

Radiologists and other physicians who used x-rays in their practice before strict health physics practices were common showed a significantly higher rate of leukemia than did their colleagues who did not use radiation. Among American radiologists, the doses associated with the increased rate of leukemia were on the order of 100 rads (1 Gy) per year. With the increased practice of health physics, the difference in leukemia rate between radiologists and other physicians has been continually decreasing.

Among the survivors of the nuclear bombings of Japan, there was a significantly greater incidence of leukemia among those who had been within 1500 meters of the hypocenter than among those who had been more than 1500 meters from ground zero at the time of the bombing. An increase in leukemia among the survivors was first seen about three years after the bombings, and the leukemia rate continued to increase until it peaked about four years later. Since this time, the rate has been steadily decreasing.

The questions regarding the leukemogenicity of low radiation doses and of the existence of a non-zero threshold dose for leukemia induction remain unanswered, and are the subject of controversy. On the basis of a few limited studies, it was inferred that as little as 1-5 rads (10-50 mGy) of x-rays could lead to leukemia. Other studies imply that a threshold dose for radiogenic leukemia is significantly higher. However, it is reasonable to infer that low level radiation at doses associated with most diagnostic x-ray procedures, with occupational exposure within the recommended limits, and with natural radiation is a very weak leukemogen, and that the attributive risk of leukemia from low level radiation is probably very small.