Abstract
CN02-02
Despite the fact that acute myelogenous leukemia (AML) is a heterogeneous disorder almost all descriptive epidemiological studies consider AML as a single entity. As a consequence, information on specific homogenous subtypes is very limited and etiological clues of homogenous AML subsets are likely to have been masked by studying AML as a whole. Characterization of AML patients by specific chromosomal aberrations divides the subgroups into true homogenous entities. Further, the different chromosomal abnormalities result in specific gene rearrangements that are often involved in the etiology and the pathogenesis of the disease. Acute promyelocytic leukemia (also called M3 by the FAB classification) is a subtype of AML in which the leukemia cells are morphologically characterized as blasts and promyelocytes In almost all cases APL is characterized by a unique cytogenetic abnormality that is a balanced reciprocal translocation between chromosomes 15 and 17, i.e. t(15;17). The breakpoint sites on chromosomes 17 and 15 are within the retinoic acid receptor α (RARα) and the PML genes, respectively. The translocation fuses most of the RARα gene from chromosome 17 with most of the PML gene at the breakpoint of chromosome 15. The fusion PML-RARα gene located on 15q+ chromosome expresses APL specific PML/ RARα transcripts in almost all cases of this disease. PML/RARα gene expression plays a crucial role in the pathogenesis of the APL. The breakpoint in the RARα gene has consistently been described in intron 2 while the PML breakpoints may occur in three different sites: intron 6 (bcr1 or long form), exon 6 (bcr2 or variable form), or intron 3 (bcr3 or short form) The distribution of the breakpoint sites in the PML gene has been reported in several studies from Europe to be approximately 50%-55% for bcr1, 8%-20% for bcr 2 and 27%-49% for bcr 3. Reports suggesting an association between different bcr breakpoint sites and clinical characteristics or response to treatment in APL patients have not been consistent The specific chromosomal abnormality involving of APL allows us to study the epidemiology of a homogeneous AML subtype, and the distinct PML/RARα molecular abnormality, may enable us to better decipher its etiologies causes. Further, APL or FAB M3 is fairly easy to recognize by morphology criteria with minimal error and therefore the translocation t(15:17) can be identified in most patients without actually performing a cytogenetic analysis. Thus large databases of AML can be used to study the epidemiology of APL and circumventing the lack of cytogenetic information. Since 1972, APL has been coded in SEER registries separately and distinctly from all other AML cases, by the morphological features of the leukemia cells. Using such databases we were able to study the epidemiology of APL focusing on the ethnic distribution of this disease. Approximately 12,000 new cases of AML are diagnosed in annually U.S. The incidence increases with age espacialy after age 55-60 the when the disease also changes its biology and becomes less responsive to treatment, Several large studies in USA and Europe approximately 10.8%, of AML patients had APL. Thus, one would estimate approximately 1,200 new case of APL per year in the U.S. Throughout the world, registries report a higher incidence of AML in males than in females. In contrast the incidence of APL for females and for males is equal. There is a relatively low incidence rate of APL in childhood and then constant rate over the rest of the life span. The stable age distribution in APL, is in contrast to the well known increase in the incidence rates of non-APL AML with. The increase rate with age of AML is believed to result from the multi-step process of accumulation of mutations during a life long exposure to environmental leukemogenic causes. In contrast, the relatively constant incidence of APL with age, could imply that less genetic mutations are required, suggesting different etiology of APL than other AML subtypes In the mid 1990s, our group noticed a high frequency of APL cases among Latino patients with AML diagnosed in Los Angeles County (LAC) - USC Medical Centre in Los Angeles. Of 80 cases of AML among patients originating in Latin America countries, 37.5% had the APL subtype as opposed to only 6.5% of 62 all other non-Latino patients with AML. In a selected sample of well characterized AML patients from the county of Los Angeles, 24.3% of Latino patients had APL compared to 8.3% of non-Latinos. This observation was subsequently confirmed by a report from Lima Peru (22%), in Mexican Mestizo patients in Puebla (20%), among Children in Mexico City (21%) and Texas. “Latinos” are not a single racial or ethnic group and could be defined in different ways. In our large population-based survey of the Los Angeles CSF database of the entire AML population of the County we found that the proportion of APL in Latinos with AML was 9.1% compared to 4.2% for non Latinos. Ruiz-Arguelles suggested that Latino populations are people speaking the Spanish language. Our definition of “Latinos” is a geographic one that includes people who originated in Latin America (Mexico, Central or South America). In a population-based study in Mexico City the AAIR of APL in children younger than 15 years was 2.21 cases per 1 million children compared to a lower rate (0.2- 0.9 cases per 1 million) among non Latino APL children in Los Angeles county (23). The reason for the high frequency of APL is Latinos is unclear and could be explained by environmental factors or genetic predisposion. We reported that the APL specific PML/RARα gene rearrangement and specific breakpoint region (bcr subtypes) in the PML gene is different in Latinos and non-Latinos. The percentage of bcr1 patients has consistently been reported as 50%-55% in several large studies, which include patients from Europe (including Spain) and from non-Latinos in the USA. In contrast, we found a statistically significant higher rate (75%) of the bcr1 subtype among APL patients originating in Latin America compared to rate published in Europe and non-Latinos in USA. This was found independently in two groups of APL patients from Latin America: one group diagnosed in the Los Angeles area, and the other group diagnosed in Lima, Peru. This finding was confirmed in Mexican Mestizo patients with 63% brc1 cases. Thus, the overrepresentation of APL among Latin American patients with AML may be accounted for by an increase in a single isoform, bcr1, rather than an equal increase in all three bcr isoforms. For unknown reasons, APL patients from Latin America may have a predilection for a break at intron 6 in the PML gene. Since the PML/RARα fusion protein gene is involved in the pathogenesis of APL, the particular breakpoint site of the PML gene (for example in intron 6 in Latinos) might be associated with an etiology that may be determined genetically. The two populations in our study and study from Mexico share a high frequency of “Mestizos”, meaning different degrees of genetic mixture of Caucasian and Indian races. Because these patients originated from places, which are far apart in the American Continent, it is suggest that the higher rate of bcr1 might be genetic than environmental. In Peru, as in several other Latin American countries, the population is composed of pure Indians, pure Europeans and mostly Mestizos. Because of the different distribution of the bcr subtypes between Europe (including Spain) and Latin America, we suggest a hypothesis that the higher rate of bcr1 APL subtype in Latin American patients might be related to a non-European genetic factor, possibly originating in Native Americans, in the ethnically mixed Latin American genetic pool. Interestingly, in a small cohort of Chinese APL patients and in a Japanese group, the bcr1 rates were the closer to Latin American patients in Los Angeles, Peru and Mexico. One could speculate that a non-European genetic factor might have migrated from the East Asia through the Behring Straits into America approximately 12,000 years ago. The treatment of APL with the combination of chemotherapy and all trans retinoic acid results in a very high complete remission rate (90%), few relapses, and very high cure rate that in several studies in Europe and USA approaches the rate of 80-90%. A major problem in treating patients with APL is the coagulopathy associated with disease; fatal bleeding prior to treatment or early after treatment had begun occurs in approximately 5% of patients which is a major cause of failure to induce a remission. Recently Ribeiro and Rego summarized the outcome of APL patients is Mexico, several Central America countries and Brazil and reported a poorer outcome. Although most patients received ATRA the CR rate in this survey was only (42%-76%) with the bleeding being the main cause of early death. The survival rate was reported in this survey to be 17%-50%. The significant improvement over the last decade in the treatment of this highly curable disease has not been yet been seen in Latin America Counties. To address this problem the American Society of Hematology has initiated a program to improve the survival of APL patients in selected developing countries. APL, with its unique and well-defined chromosomal aberration and PML/RARα gene rearrangement, provides an excellent model to study a homogenous subtype of AML. APL has distinct epidemiological characteristics, namely a relative constant incidence with age, equal incidence in males and females, and higher frequency among patients originating in Mexico, Central and South America. Future molecular biology investigations should explore the relationship between epidemiological observation and the specific PML/RARα gene rearrangement.
First AACR International Conference on the Science of Cancer Health Disparities-- Nov 27-30, 2007; Atlanta, GA