Causes and Risk Factors
Gender and Age
Leukemia does not affect all populations equally. For instance, men are more likely to develop leukemia than women, and older people are typically at higher risk than younger people. ALL is an exception to this trend (ie, ALL is more common in children) (Siegel 2011; Siegel 2012; Siegel 2013). Leukemia accounts for almost one-third of all cancers diagnosed in children from birth to 14 years, and over 75% of these childhood leukemias are ALL. Leukemia is the leading cause of cancer death among men under age 40 (Siegel 2013).
Genetics and Family History
There is strong evidence for a genetic component to some types of leukemia. People with at least one affected sibling with a hematological (blood-related) cancer have a 2.3 times increased risk of developing leukemia. Individuals reporting at least one sibling with leukemia showed three times the risk of developing CLL (Pottern 1991). In a study on twins, those with an identical twin affected by leukemia had a greatly increased chance of developing leukemia themselves (Kadan-Lottick 2006). Family history of other types of cancer may be a risk factor for adult leukemia as well (Poole 1999; Wang, Lin 2012).
Certain genetic abnormalities, such as Down syndrome, are associated with leukemia. Studies suggest that children with Down syndrome have an almost 20-fold greater risk of developing leukemia than the general population. In this population, the highest incidence of leukemia is observed in children less than 5 years of age (Ross 2005).
Among certain types of leukemia, patients with specific genetic abnormalities have an increased risk of developing resistance to therapy, and possibly a greater chance of relapsing after remission (Meijerink 2009; Estey 2010; Medeiros 2010). One example of this is a monosomal karyotype, a chromosomal abnormality that is a strong predictor of drug resistance and poor prognosis in patients with AML (Estey 2010; Medeiros 2010).
Alterations in specific genes play a role in some types of leukemia as well. For example, the FLT3 gene is important for normal growth and development of blood cell precursors. The protein encoded by this gene is not normally present in large amounts in mature blood cells (Karsunky 2003). However, mutations in the FLT3 gene are known to be common in AML, and FLT3 mutations are a strong predictor of poor prognosis in AML (Kottaridis 2001; Gilliland 2002).
Patients with certain pre-existing blood disorders may be at increased risk of developing some forms of leukemia. For instance, there is a higher risk of developing AML in people with chronic myeloproliferative disorders including polycythemia vera, essential thrombocythemia, and idiopathic myelofibrosis. There is additional risk when treatment for these conditions includes some types of chemotherapy or radiation (ACS 2013a).
Smoking cigarettes increases the risk of developing leukemia, particularly adult AML (Brownson 1993; Thomas 2004; Musselman 2013). In one study, the risk of leukemia increased according to the number of cigarettes smoked per day (Brownson 1993). Even children exposed to parental cigarette smoke prenatally or after birth appear to have an increased risk of childhood ALL (John 1991; Chang 2006). However, a 2013 study revealed encouraging evidence that leukemia risk decreases with increased time after quitting, and in long-term quitters (30 years or more) the risk was comparable to that of non-smokers (Musselman 2013).
Exposure to certain chemicals has been found to elevate the risk of developing leukemia. For example, long-term exposure to benzene, a constituent of crude oil and known cancer-causing agent, increases the risk of leukemia (Snyder 2012; Li 2014; Yin 1996; Savitz 1997; Hayes 2001; ACS 2013b). Benzene is used as a starting material to make a wide variety of substances including plastics and pesticides. It is a common chemical in the environment throughout the United States, with high levels in the vicinity of gasoline stations and some industrial facilities, vehicle exhaust, and secondhand tobacco smoke (Brugnone 1997. Therefore, people who work in chemical plants, oil refineries, and gasoline-related industries may be exposed to high benzene concentrations (ACS 2013b). A major source of benzene exposure is tobacco smoke. Estimates suggest that benzene in cigarettes is responsible for about one-third of smoking-induced AML (Korte 2000; Vineis 2004). Formaldehyde is another environmental chemical potentially linked to leukemia. Formaldehyde is generated by automobile engines, is a component of tobacco smoke, and is released from household products, including furniture, particleboard, plywood, and carpeting (Zhang 2009). The association between formaldehyde exposure and leukemia risk is controversial, however, as some studies support the link (Schwilk 2010) while others do not (Checkoway 2012; Gentry 2013). Agent Orange, a chemical defoliant to which many soldiers were exposed during the Vietnam War, has also been associated with increased leukemia risk (Yi 2013; Baumann Kreuziger 2014). Parental pesticide exposure near the time of conception or during pregnancy has been associated with childhood leukemia, as has childhood exposure (Turner 2010; Turner 2011; Ferreira 2013). Among adults, living on or near a farm has been linked with a greater risk of developing or dying from leukemia, possibly as a consequence of increased agricultural pesticide exposure (Viel 1991; Jones 2014). Exposure to multiple pesticides may exacerbate the risk of malignancy, as experimental evidence shows that mixtures of pesticides, at low concentrations, can damage DNA to a similar extent as higher concentrations of single pesticides alone (Das 2007).
Exposure to high-energy (ionizing) radiation (eg, atomic bomb explosions) is linked to leukemia. For example, there was a dramatic increase in leukemia risk among Hiroshima and Nagasaki atomic bomb survivors soon after the bombings (Hsu 2013; Preston 1994); and in one study, the excess risk persisted, especially for AML, even 55 years after the bombings (Hsu 2013). Also, exposure to lower doses of radiation from post-Chernobyl cleanup work has been associated with a significant increase in the risk of leukemia (Zablotska 2013). Some evidence suggests that low-dose radiation, such as from diagnostic X-rays, may be associated with increased leukemia risk as well. In one study, children who had undergone X-ray examinations were at increased risk of childhood leukemia (Shih 2014). Another study found similar results: children who had been exposed to post-natal X-ray examinations for diagnostic purposes had an increased risk of childhood ALL (Bartley 2010). Computed tomography (CT) scans also appear to increase cancer and leukemia risk in children, though it is thought that the diagnostic benefits generally outweigh the risks (Pearce 2012). Older CT scanning technology resulted in higher radiation exposure, though current CT scans are believed to still carry some degree of risk (Mathews 2013). Thus, non-radiative diagnostic measures in children, when appropriate, are considered preferable (Miglioretti 2013; Knusli 2013; ARSPI 2014).
The possibility of a relationship between living in close proximity to high-voltage electricity lines and the risk of childhood leukemia has been studied for decades (Washburn 1994). This remains a contentious issue (Magana Torres 2013; Clavel 2013); some studies failed to find a relationship (Pedersen 2014), while other studies found an association ranging from significant (Washburn 1994; Sohrabi 2010) to not achieving statistical significance (Sermage-Faure 2013). Those that found a relationship suggested risk increased markedly with closer proximity to such power lines (Sidaway 2013; Roosli 2013).
Previous Cancer Treatment
Aggressive chemotherapy and radiation therapy can improve outcomes for many cancer patients. Unfortunately, high-dosage treatment regimens also significantly increase the risk of development of subsequent leukemia (Huh 2013; Pedersen-Bjergaard 2000; Morton 2013; Kaplan 2011).
Therapy-related myelodysplastic syndromes are a major complication among patients treated for previous blood-related malignancies or solid tumors (Leone 2007; Leone 2010; Zompi 2002). A 2013 study examined the records of over 426 000 US cancer patients who were treated with chemotherapy for a primary tumor between 1975 and 2008. Among this group, a subsequent diagnosis of AML was 4.7 times more common than expected in the general population (Morton 2013). A similar study examined the records of over 5700 breast cancer patients diagnosed between 1990 and 2005. A 10.9-fold increased risk of MDS was found in breast cancer patients under age 65, and a 5.3-fold increased risk of AML was noted in the same population. The risk was higher for those who received a combination of chemotherapy and radiation compared to those who received chemotherapy or radiation alone (Kaplan 2011). Studies indicate that childhood cancer survivors are at an increased risk of developing malignancies during their later years compared to peers who did not have childhood cancer. In one study, children who survived more than five years after having AML had 3.9-times greater risk of a secondary cancer, and those who survived more than five years after having ALL had a 4.3-times greater risk compared with the general population. This study followed more than 4800 childhood cancer survivors for 14.5 years (Perkins 2013; Joh 2011). Overall, the risk of developing leukemia is increased as a result of cancer therapy. Thus, lifelong surveillance is recommended for cancer survivors (Vega-Stromberg 2003).
Chemotherapeutic drugs such as alkylating agents, nitrosoureas, procarbazine (Matulane), and topoisomerase II inhibitors have considerable potential to increase the risk of leukemia. Exposure to large, cumulative doses of alkylating agents is a prominent risk factor for leukemia secondary to chemotherapy (Leone 2010; Pedersen-Bjergaard 2000). In addition, some supportive agents administered along with rigorous chemotherapy have also been found to increase the risk of leukemia. For example, colony stimulating factor (CSF) use among elderly patients with non-Hodgkin’s lymphoma undergoing chemotherapy has been associated with an increased risk of developing leukemia (Gruschkus 2010).
Human T-cell lymphotropic virus type 1 (HTLV-1), the first retrovirus shown to cause human malignancy, was identified as the causative agent of adult T-cell leukemia-lymphoma (ATLL) (Yoshida 1982; Beltran 2009; Zane 2014). HTLV-1 mainly affects specialized immune T cells (CD4+ T cells), causing infected T cells to transform into cancerous cells (Satou 2013). In addition, multiple studies have shown a correlation between maternal and childhood infections and subsequent risk of developing childhood leukemia. Some studies have also found a significantly increased risk of childhood leukemia in children whose mothers experienced infections during pregnancy (Sadrzadeh 2012). For example, children whose mothers experienced a reactivation of the Epstein-Barr virus during pregnancy were found to have an almost three-fold greater risk of developing ALL in one study (Lehtinen 2003).