18

DESCRIPTIVE EPIDEMIOLOGY OF CHRONIC LYMPHOCYTIC LEUKEMIA (CLL)   GE MARTI, FD GROVES AND MS LINET   INTRODUCTION  
    A previous review of the epidemiology of CLL discussed methodologic issues, demographics, genetic and occupational factors and second neoplasms (1). Environmental and occupational exposures are discussed elsewhere in this volume and second neoplasms subsequent to CLL have recently been reviewed (2). This brief review for the proceedings volume will focus on descriptive epidemiology (ie, variation in CLL incidence rates with age, sex, race, time, and geographic location).
  DEMOGRAPHICS  
    Internationally and within the United States, CLL accounts for approximately 30% of all leukemias. Reported CLL rates differ by more than 30 fold among populations, the greatest variation among all types of leukemia (3).  Rates (See Figure 1) are highest in Israel, Australia, Europe (western, northern, and eastern) and North America (north central US and the contiguous central provinces of Canada). Lower rates are seen in Latin America and the Caribbean and very low rates of CLL are seen in China where it is rare and in India, Japan and Singapore.
  MALE/FEMALE RATIO  
    There is a marked male predominance in CLL. Interestingly, Linet and Cartwright note that the male/female ratio was higher earlier in this century (2.5-3.0) while more recent studies report lower values (1.6-1.9). And just as there is an incidence variation of CLL in various populations, there is variation in the male/female ratio. It is 4.7 in Australia and 1.0 in Colombia (See Figure 1). The highest rates for males are in Canada, Denmark and the other Scandinavian countries (3). While a female predominance was noted in young adult Nigerians (4,5,6), this may be due to chance since a male predominance and similar age distribution as in western populations was noted in a hospital series from Kenya (7). Females in Cali, Colombia and Warsaw, Poland have very low rates (3).
    RACE AND ETHNICITY  
    The age-specific incidence of CLL rises steeply after age 30 in both US whites and blacks (see Figure 2). The marked variation noted worldwide is also seen among United States racial and ethnic groups (see Table 1). The highest rates are among whites, high though slightly lower rates in blacks, and very low rates among others (this residual category combines American Indians, Alaska Natives, Chinese, Japanese, Filipinos, Hawaiians, Koreans, Asian Indians, Pakistanis, and Pacific Islanders). Small differences are seen between US whites and blacks of the same sex. However, the CLL rate in black African populations may be lower than among most Caucasian populations, although the absence of high-quality population denominator data restricts comparison of incidence rates (3). A number of studies have also suggested that there is an increased incidence of CLL among Jews, particularly those of Russian or Eastern European origin compared to those of Asian or north African origin (3,8).
  MAPPING  
    Age-Adjusted incidence rates from the US SEER Program show a marked north-south gradient, with the highest rates reported in the states of Iowa and Connecticut and the Seattle and Detroit metropolitan areas; rates in the states of Hawaii and New Mexico and the Atlanta and San Francisco metropolitan areas are lower. The most comprehensive large scale study of the incidence of hematological malignancies was conducted in England and Wales by Cartwright and colleagues and covers the five year period 1984-1988 (9). Their Data Collection Study (DCS) includes 3,340 cases of chronic lymphoid leukemia.  Under this heading, the following percentages were noted: CLL/PLL (prolymphocytic leukemia), 90%; mycosis fungoides (MF), 5%; Sezary's syndrome (SS), 0.3%; hairy cell leukemia (HCL), 5% and large granular lymphocytosis (LGL), 0.3%. The heading CLL/PLL contained common B-CLL, T-CLL, B-PLL and T-PLL, although the latter three entities are quite rare. The DCS data confirm the steep rise in the age-specific incidence of B-CLL with advancing age in both sexes. The incidence rates over the five year period were stable in the UK as also seen in the US in contradistinction to NHL, which has been rapidly increasing for several decades. Within England and Wales, the maps of CLL and low-grade NHL showed a similar geographic distribution.  Reasons for the geographic variation in rates in the UK are unknown.
      CELL TYPE  
    The SEER database allows for phenotype coding (B-cell, T-cell, Null-cell, other, or unknown) on all lymphomas and leukemias. Over the most recent ten-year period (1983-1992) for which SEER data are available, an increasing proportion of leukemia and lymphoma cases have included phenotype codes.  By 1992, almost 40% of non-Hodgkin's lymphomas had phenotype codes, and the majority of these were of the B-cell type. The same pattern was observed for acute lymphoid leukemia. Only 13% of chronic lymphocytic leukemias had phenotype codes by 1992, and again the vast majority were of the B-cell type (Figure 3). Since most leukemia and lymphoma cases in the SEER database still lack phenotype codes, it would be inappropriate to attempt to calculate incidence rates separately by phenotype at the present time. However, if present trends continue, such calculations may be possible in the future.
    SUMMARY  
    The marked variation in the incidence of CLL both within and between countries is striking in comparison to the other leukemias. There is a gradient from the east to the west hemispherically and from the south to the north in the US. Therefore the highest incidence rates are reported in the north central US and the contiguous central provinces of Canada. The variation in the male/female ratio is of equal interest. These observations suggest that there is at least one gene involved in CLL that is modified by gender, ethnic origin and environment. All of these phenotypes must be genetically based. In future studies, these epidemiologic findings need to be correlated with the differential diagnosis of B-CLL and flow cytometric patterns. Comparative molecular studies will be required to identify candidate genes associated with these epidemiologic findings.
  REFERENCES  
  1. Linet MS and Blattner WA: The Epidemiology of Chronic Lymphcytic Leukemia. In: Polliack
     A and Catovsky D (eds.): Chronic Lymphocytic Leukemia Harwood Academic Publishers,
     New York, 1988, pages 11-32.
  2. Travis LB, Curtis RE, and Hankey BF: Second Cancers in Patients with Chronic
      Lymphocytic Leukemia. J Natl Cancer Inst 1992; 84:1422-1427.
  3. Parkin DM; Muir CS; Whelan SL; et al.: Cancer Incidence in Five Continents, Volume Six.
      Lyon, IARC Scientific Publication Number 120, 1992.
  4. Fleming AF: Chronic lymphocytic leukemia in tropical Africa: A review. Leukemia and
      Lymphoma 1990; 1:169-173.
  5. Okpala I: Contrasting sex distribution of chronic lymphocytic leukemia and well-differentiated
      diffuse lymphocytic lymphoma in Ibadan, Nigeria. Eur J Cancer 1990; 26:1105.
  6. Williams CKO: Neoplastic diseases of the hematopoietic system in Ibadan: Preliminary
      report of a prospective study. Afr J Med Sci 1985; 14:89-94.
  7. Oloo AJ and Ogada TA: Chronic lymphocytic leukemia (CLL): Clinical studies at Kenyatta
      National Hospital (KNH). E Afr Med J 1984; 61:797-801.
  8. Steinetz R, Parkin DM, Young JL, Bieber CA, Katz L: Cancer Incidence in Jewish Migrants
      to Israel, 1961-1981. Lyon, IARC Scientific Publications No. 98, 1989, p.238.
  9. Cartwright RA, Alexander FE, McKinney PA, and Ricketts TJ: Leukaemia and Lymphoma:
      an atlas of distribution within areas of England and Wales 1984-1988. Complied by the
      Leukaemia Research Fund Centre for Clinical Epidemiology at the University of Leeds.
      Published by the Leu-kaemia Research Fund, 43 Great Ormond Street, London WC1N 3JJ.
      Pages 58-72.
   Return To Contents