Take, for example, a relatively rare form of leukemia known as AML (acute myeloid leukemia) that predominantly affects middle-age and elderly adults. Based on data from NCI and others recently released in the Annual Report to the Nation, an estimated 14,000 people will develop AML in the United States in 2013, and an estimated 10,000 people will die from it. Although life expectancy for AML survivors has increased slowly over the past decade, the improvement is predominantly because of advances in supportive care, not in the therapies that patients received.
Primary AML, or that which develops before any other diagnosis of cancer, is the form of the disease that is most common and does not display a cohesive set of genetic mutations that contribute to its initiation or progression. Secondary AML is related primarily to earlier treatment, usually chemotherapy. The incidence rate for all cases of AML has been about 3.5 new cases per 100,000 per year of the U.S. population since the mid-1970s, even when secondary AML is included in the rate. As seen in the graph below, the incidence rate for primary AML has been even more level and hovered around 3 per 100,000 new cases per year since the early 1970s.
It should be noted however that when the numbers are parsed, a comprehensive report covering the period 2001-2008 showed strong increases in incidence with increasing age for most AML subtypes,. The data also showed statistically significant differences in the incidence of certain AML subtypes by race. So a trend that looks straightforward on the surface may display a number of quirks based on a different slicing of the data.
In part because of these somewhat divergent trends, experts have been working for years to understand the genetic basis of this cancer type, and the risk of developing treatment-related AML after chemotherapy. Researchers know that certain chemotherapy treatments run a high risk of predisposing patients to a second leukemia at a later age; however those risks are necessary to treat the current cancer. There are also some preliminary studies that suggest there is a genetic susceptibility to developing AML later on, based on earlier treatments for a patient’s initial cancer.
In a study reported in Blood in February 2013, the risk of secondary AML among patients initially treated for non-Hodgkin lymphoma (NHL) was found to increase steadily during the subsequent few decades. Over the same time period, the researchers observed declining risk among patients treated for ovarian cancer, myeloma, and possibly lung cancer. The authors indicated that the increased risk among NHL survivors could be due to prolonged survival in recent years for some lymphoma subtypes that are associated with multiple courses of chemotherapy. The decreased risk among patients with ovarian cancer was consistent with a shift from use of an alkylating agent called melphalan to platinum-based chemotherapy in the early 1980s.
Much of the data that make possible hypotheses of a cancer’s origin come from NCI’s Surveillance, Epidemiology, and End Results (SEER) Program, which collects cancer incidence, mortality and survival data from population-based cancer registries that today cover approximately 28 percent of the U.S. population. This information is incorporated into a web resource called the Cancer Statistics Review. The SEER Program has been collecting data since 1973, making it a unique resource for researchers who assess long-term data trends, such as those related to AML. In addition to collecting data, SEER compiles and reports death rates and trends for the entire United States based on data from the National Center for Health Statistics of the Centers for Disease Control and Prevention (CDC).
SEER is also unique in that it if statisticians detect changes and trends in a particular cancer, they can respond to those changes by collecting new types of information to help better assess the situation. For example, starting in 2000, SEER started collecting information on MDS (Myelodysplastic Syndromes) and CMD (Chronic Myeloproliferative Disorders) that can progress to AML. This brings up a problem, however: if the data is collected as MDS or CMD that then progressed to AML, the cases would not be recorded as AML because it was not a new tumor. So there was an additional change in data assessment of 2010 cases in order to collect AML information even if it was a progression from MDS or CMD. SEER experts will carefully track these cases to see if there is an increase next year with the change to collect AML even if preceded by MDS or CMD. Of particular note is the fact that because of the limited number of cases, this is a hard cancer to capture statistically.
The continued refinement and broadened coverage of cancer cases nationwide should only aid in helping geneticists and others glean important insights from these statistics, much in the way AML researchers have used this data in their search for greater understanding of a still rather mystifying and complex cancer.