Gregory T. Armstrong, MD, MSCE
Over the past 30 years, survivors of childhood cancers have become less likely to die of illnesses caused by their treatment, such as new malignancies or cardiac or lung disease. The improvement is credited, in part, to better cancer treatment and follow-up, including the reduced use of radiotherapy and anthracyclines.
The findings come from an analysis of more than 34,000 participants in the federally funded Childhood Cancer Survivor Study (CCSS), which found that among those who have survived at least 5 years all-cause mortality at 15 years after diagnosis dropped from 12.4% to 6%.
The results were announced in a press briefing May 31 during the ASCO 2015 Annual Meeting.
“To go back to the 1960s, less than 30% to 40% of children would survive cancer; currently over 83% of children (with the disease) will become 5-year survivors. It’s estimated that, at 2013, there were over 400,000 survivors in the population, and that by 2020 there will be over half a million,” said lead study author Gregory T. Armstrong, MD, MSCE, a pediatric oncologist at St. Jude Children’s Research Hospital.
“However, these 5-year survivors are still at risk for late effects and late mortality. Thus, strategies to reduce late effects by reducing treatment intensity have been used for several decades in an attempt to lower mortality rates and improve quality of life.”
According to previous research, up to 18% of 5-year survivors of childhood cancer died within 30 years of diagnosis, Armstrong noted during his presentation. Such deaths are attributable to progression or recurrence of the primary cancer, external causes such as accidents or suicide, or other health-related causes—primarily, the late effects of cancer therapy, ASCO explained in a statement.
Investigators used the National Death Index, a record of death information kept by state vital statistics offices, to determine which patients in the cohort had died and when. All participating patients were diagnosed between 1970 and 1999, were younger than 21 years of age at diagnosis, and were followed for a median 21 years. Their cancers included leukemia, lymphoma, malignancies of the central nervous system, Wilms tumor, neuroblastoma, and soft-tissue or bone sarcoma, Armstrong said.
The study found that 3958 patients (12%) died during their follow-up period, with 1618 (41%) of those deaths due to other health-related causes, including the late effects of cancer therapy—in particular, secondary or recurrent cancers.
All-cause mortality dropped by half over the study period—12.4% of patients diagnosed in the early 1970s died within 15 years of diagnosis, compared with only 6% of those diagnosed in the early 1990s. In those who survived at least 5 years, the cumulative incidence of deaths due to other health-related causes—including late effects such as second cancers or heart or lung disease—dropped with each decade. For those diagnosed in the 1970s, the cumulative incidence of deaths from other health-related causes was 3.1%; for those diagnosed in the 1980s, it was 2.4%, and for those diagnosed in the 1990s, it was 1.9%, Armstrong said.
Deaths from subsequent neoplasms in 5-year survivors of childhood cancers dropped in frequency from 1.8% in the early 1970s to 1% in the early 1990s; deaths from cardiac late effects fell from 0.5% to 0.1% in those same time periods, respectively, and deaths from pulmonary late effects from 0.4% to 0.1%, he pointed out.
The improvement was most prevalent among survivors of Wilms tumor, Hodgkin lymphoma, and acute lymphoblastic leukemia (ALL). Cardiac deaths significantly decreased among survivors of all three cancers, although deaths due to secondary cancer decreased among Wilms tumor survivors only. In ALL, deaths at 15 years after diagnosis due to late effects were reduced from 2.8% in the 1970s to 1.9% in the 1990s; in Hodgkin lymphoma, they dropped from 4.2% to 2.1% in those time periods, respectively, and in Wilms tumor, they dropped from 2.2% to 0.4% in those periods, Armstrong said.
Doctors have helped to drive these results by gradually refining treatment through reductions in the intensity of therapy for many pediatric cancers, even as the effectiveness of treatment has risen, Armstrong said. For example, in the 1970s, 86% of patients with ALL received cranial radiotherapy, compared with only 22% in the 1990s, he and coauthors wrote. Radiotherapy rates have also been reduced among patients with Hodgkin lymphoma (from 96% to 77%) and Wilms tumor (from 77% to 49%). Furthermore, Armstrong said, patients who develop these three diseases in childhood are now being exposed to less anthracycline.