Second malignancy after Hodgkin disease treated with radiation therapy with or without chemotherapy: long-term risks and risk factors

Abstract

The excess risk of second malignancy after Hodgkin disease is an increasing problem. In light of the long-term data, guidelines for follow-up of survivors of Hodgkin disease need to be redefined. In this study we attempt to analyze the longterm risks and temporal trends, identify patient-and treatment-related risk factors, and determine the prognosis of patients who develop a second malignancy after radiation treatment with or without chemotherapy for Hodgkin disease. Among 1319 patients with clinical stage I-IV Hodgkin disease, 181 second malignancies and 18 third malignancies were observed. With a median follow-up of 12 years, the relative risk (RR) and absolute excess risk of second malignancy were 4.6 and 89.3/10 000 person-years. The RR was significantly higher with combined chemotherapy and radiation therapy (6.1) than with radiation therapy alone (4.0, P ‫؍‬ .015). The risk increased with increasing radiation field size (P ‫؍‬ .03) in patients who received combined modality therapy, and with time after Hodgkin disease. After 15 and 20 years, there was a 2.3% and 4.0% excess risk of second malignancy per person per year. The 5-year survival after development of a second malignancy was 38.1%, with the worst prognosis seen after acute leukemia and lung cancer. The excess risk of second malignancy after Hodgkin disease continues to be increased after 15 to 20 years, and there does not appear to be a plateau. Our analysis suggests that the risk may be reduced with smaller radiation fields, as are used in current trials of abbreviated chemotherapy and limitedfield radiation therapy.

Second primary breast cancer after Hodgkin's disease

“…Acute leukaemia frequently occurs in the first decade after treatment, mainly as a consequence of chemotherapy regimens that included an alkylating agent (Tucker et al, 1988;Kaldor et al, 1990;Swerdlow et al, 2000). In the long term, however, the absolute excess risks of a second solid cancer are higher (Swerdlow et al, 2000;Ng et al, 2002b; and are linked mainly with radiotherapy, although for some sites, such as lung cancer, there may be a substantial risk also from chemotherapy (Swerdlow et al, 2000;Travis et al, 2002). Recent reports from The Netherlands and from an international consortium of cancer registries have refined our understanding of how therapy may affect breast cancer risk.…”

“…For patients treated under age 21 years relative risks have generally been around 15 -25 (Hancock et al, 1993;Sankila et al, 1996;Wolden et al, 1998;Metayer et al, 2000;van Leeuwen et al, 2000;Dores et al, 2002), although they have been less raised (Neglia et al, 2001) or more raised (Mauch et al, 1996;Bhatia et al, 1996b;Aisenberg et al, 1997;Ng et al, 2002b) than this in a few studies; absolute excess risks have generally been of the order of 20 -40 per 10 000 per annum. Relative risks have been greater for patients treated at ages 10 -16 than at younger or older ages (Hancock et al, 1993;Sankila et al, 1996;Bhatia et al, 1996b;Metayer et al, 2000).…”

“…In adults, at least, the induction period before breast cancer is observed is long, with significantly raised risks appearing only at 10 -14 (Swerdlow et al, 2000;Dores et al, 2002;Ng et al, 2002b) or 15 -19 (Hancock et al, 1993van Leeuwen et al, 2000) years after first treatment. After childhood or adolescent treatment, however, induction periods appear to be shorter (Bhatia et al, 1996b;Sankila et al, 1996;Travis et al, 1996;Metayer et al, 2000), indeed as short as within the first 5 years in one study (Bhatia et al, 1996b).…”

“…Similarly, estimates of cumulative risk of breast cancer after treatment at young ages have at the extreme been estimated at 42% at 30 years of follow-up for women treated under age 16 years in a study with incomplete follow-up (Bhatia et al, 1996a), but studies with highquality follow-up have found cumulative risks of 12% at 30 years after treatment in the Nordic countries (Sankila et al, 1996) and 16% at 25 years in the Netherlands for patients treated under age 21 years. In median 18-year follow-up of over 500 women treated at ages under 41 years in the Dutch cohort, only five deaths occurred from breast cancer , and among 161 deaths in long-term follow-up of 494 women treated under age 51 years in the US, four were from breast cancer (Ng et al, 2002a) Risks of second primary breast cancer are also raised after treatment for Hodgkin's disease in young adulthood, although relative risks are lower than those for childhood and adolescent treatment (Hancock et al, 1993;van Leeuwen et al, 2000;Ng et al, 2002b), reflecting the greater background rates at older ages. However, a lesser relative risk does not necessarily imply a lesser absolute excess, and indeed van in the Netherlands and Hancock et al (1993)and Ng et al (2002b) in the US found absolute excess risks fairly similar for patients treated at ages 20 -29 years to those for patients treated under age 20 years.…”

“…In median 18-year follow-up of over 500 women treated at ages under 41 years in the Dutch cohort, only five deaths occurred from breast cancer , and among 161 deaths in long-term follow-up of 494 women treated under age 51 years in the US, four were from breast cancer (Ng et al, 2002a) Risks of second primary breast cancer are also raised after treatment for Hodgkin's disease in young adulthood, although relative risks are lower than those for childhood and adolescent treatment (Hancock et al, 1993;van Leeuwen et al, 2000;Ng et al, 2002b), reflecting the greater background rates at older ages. However, a lesser relative risk does not necessarily imply a lesser absolute excess, and indeed van in the Netherlands and Hancock et al (1993)and Ng et al (2002b) in the US found absolute excess risks fairly similar for patients treated at ages 20 -29 years to those for patients treated under age 20 years. For treatment at ages 30 years and above, most studies have found no raised risk (Hancock et al, 1993;Aisenberg et al, 1997;Swerdlow et al, 2000), although in a large Dutch cohort there was a relative risk of 2.4 (0.9 -5.2) for patients treated at ages 31 -39 years and in a US cohort there were relative risks of 3.7 (1.0 -9.5) for women treated at 30 -35 years and 3.4 (0.4 -12.1) for those treated at 36 -40 years) (Ng et al, 2002b) It thus seems possible that the raised risks extend a little beyond age 30 years, but it is unclear how far.…”