Document wDnLzQxmLzbMpnN4bvyZyEEXD

DAN S. BORNE PRESIDENT 7><r - "rn~ e LOUISIANA CHEMICAL ASSOCIATION fid-mu- June 1, 1990 MEMORANDUM TO: FROM: SUBJECT: SCIENTIFIC ADVISORY COUNCIL Edward J. Flynn Health Affairs Coordinator Cancer in Louisiana articles The April 1990 issue of the Journal of LA State Medical Society published these three articles on cancer in Louisiana. I am sending them to you FYI. EJF:bg Enclosures (3) cc: M. Currier B. Withers W. Hulon 20855001 SUITE 2040. ONE AMERICAN PLACE P.0. BOX 1180 BFG07482 BATON ROUGE. LOUISIANA 70821 (504) 344-2609 OF THE LOUISIANA STATE MEDICAL SOCIETY Cancer in Louisiana April 1990 208550U2 INCLUDING: pediatric cancer in new Orleans cancer mortality in the 30s FORETELLS CANCER OF THE 80s IN LOUISIANA BFG07483 CANCER MORTALITY IN THE 30s FORETELLS CANCER OF THE 80s IN LOUISIANA JEAN FIKE CRAIG, MSHyg; VIVIEN W. CHEN, PhD; ELIZABETH T.H. FONTHAM, DrPH; THOMAS BALLINGER, MA; PELAYO CORREA, MD Examination of secular trends suggests that the high cancer mortality rates of the 1980s could have been foretold by the excessive rates in the 1930s in Louisiana. Those same sites which were excessive in the 1930s, reflecting exposure to risk factors in the early 1900s, still predominate in the 1980s. Because of the poor survival associated with lung cancer and the increase in smoking, the increase in lung cancer had the greatest impact on the mortality trend in the all sites combined category. The stability of the cancer problem and the lack of progress in reducing cancer over the last 50 years is seen as a harbinger of the cancer problem over the next 50 years. he current high cancer rates in Louisiana are not Tunprecedented. They have been documented since the 1930s.1 This fact is especially noteworthy since cancer has a latency period of decades suggesting exposure to relevant risk factors in the early 1900s. In 1930-1932 all sex-race groups in Louisiana showed excessive cancer mortality rates for the oral cavity and bladder when compared to the United States. Males had excessive rates for cancers of the lung and pancreas while higher stomach cancer rates were noted in blacks (Table 1). Fifty years later, excess in those same cancers predominates in Louisiana. Furthermore, the rapid and vast increase in lung can cer, which has very poor survival, has resulted in a higher rate than the national average in the all sites combined category (Table 2). CANCERS OF THE RESPIRATORY TRACT Currently whites still experience excessive mortality rates for cancer of the lung and the oral cavity when compared to those of the United States. Cancers of the respiratory system cause the greatest loss of life JOURNAL VOL 142 APRIL 43 20855003 BFG07484 TABLE 1 AVERAGE ANNUAL AGE-ADJUSTED (US 1930) MORTALITY RATES FOR SELECTED SITES, BY SITES, BY SEX AND RACE IN UNITED STATES AND LOUISIANA, 1930-1932 All Sites Buccal cavity and Pharynx Stomach/Duodenum Intestine Rectum and Anus Pancreas Larynx Lung and Pleura Breast Uterus (Corpus and Cervix) Prostate Bladder Kidney Brain WM WF BM US 89.8 110.6 52.4 La 88.8 99.0 63.0 us 6.3 1.4 2.9 La 9.7 1.7 4.2 us 24.9 17.9 17.6 La 21.7 16.0 21.2 us 8.7 11.4 3.4 La 4.9 7.3 3.1 us 5.0 4.3 2.4 La 3.2 4.6 2.9 us 2.9 2.5 1.6 La 3.7 1.5 2.5 us 1.5 0.2 0.5 La 2.6 -- 0.6 us 3.2 1.8 1.4 La 4.0 1.6 1.9 us -- 19.4 -- La ~ 14.7 -- us -- 23.1 -- La -- 24.9 -- us 8.4 -- 5.7 La 6.7 8.1 us 4.3 2.1 2.3 La 5.1 2.8 2.5 us 1.6 1.3 0.8 La 1.3 1.7 1.1 us 0.9 0.7 0.3 La 0.4 0.3 0.2 BF 101.0 108.2 1.8 2.7 15.4 17.1 4.1 2.9 3.8 5.0 1.0 0.3 0.1 0.1 0.8 0.6 14.9 15.4 41.6 46.9 -- -- 1.5 2.1 0.7 0.6 0.2 0.2 US -- Excludes the state of Texas Source: Cancer Mortality in the United States, 1930-32. By M. Gover, 1940. Tables 2, 3; B and C. Rates in Tables 1 & 2 are age-adjusted to different standard populations (US 1930 & US 1970). Therefore the two tables cannot directly compared to each other. trOOSSSO in Louisiana.2 In 1987, cancer of the respiratory sys tem was the cause of death for 2,512 people in Lou isiana. All urban areas in Louisiana except Monroe have respiratory cancer mortality rates above the US average (Fig 1), in contrast to colorectal cancer which exceeds the US rate in only one metropolitan area (Fig 2) even though colorectal cancer is the second leading cause of cancer death in Louisiana. Rates for respir- 44 JOURNAL VOL 142 APRIL atory cancer in New Orleans are more than 307c higher than the US rate while the Lake Charles rate is more than 257c higher. The magnitude of excess in lung cancer and its relationship with other cancers can be graphically seen in Fig 3. It has been estimated that lung cancer ac counts for 857c of the excess of cancer in Louisiana when compared to the United States.-'1 Since the 1950s BFG07485 TABLE 2 AVERAGE ANNUAL AGE-ADJUSTED (US 1970) MORTALITY RATES FOR SELECTED SITES. BY SEX AND RACE IN UNITED STATES AND LOUISIANA. 1978-1981 WM WF BM BF All Siles Oral cavity and Pharynx Stomach Colon Rectum Pancreas Larynx Lung Breast Cervix Uteri Corpus Uteri Prostate Bladder Kidney Brain Leukemias US 209.5 132.7 286.7 La 227.2 126.8 274.5 us 5.2 1.9 10.0 La 6.4 2.0 8.8 us 7.7 3.6 14.8 La 6.5 3.4 17.8 us 21.1 16.0 20.6 La 19.1 14.3 15.4 us 4.7 2.7 4.5 La 3.9 2.0 3.9 Us 10.5 6.8 13.5 La 11.4 7.3 13.8 us 2.6 0.4 4.9 La 2.6 0.3 4.5 us 69.3 20.2 91.4 La 83.1 22.1 88.4 us ____ 26.6 -- La -- 22.1 -- us ____ La -- us ____ La -- 3.2 -- 3.1 -- 2.0 1.3 -- -- us 21.0 ___ 43.9 La 21.1 -- 36.0 us 6.9 1.9 5.4 La 6.4 2.1 5.6 us 4.6 2.0 3.8 La 5.1 2.3 2.9 us 5.0 3.4 2.9 La 4.0 3.2 2.6 us 9.0 5.2 7.4 La .9,6 5.8 8.4 152.9 153.8 2.4 2.2 6.6 7.9 17.6 15.7 2.8 2.5 9.2 8.2 0.7 0.8 20.1 20.1 26.3 24.9 8.8 8.6 2.9 2.6 -- -- 2.6 2.7 1.7 2.4 1.9 1.9 4.6 5.3 Rates in Tables 1 & 2 are age-adjusted to different standard populations (US 1930 & US 1970). Therefore the two tables cannot be directly compared to each other. Louisiana has had the highest lung cancer death rate for white males in the whole nation.4 Alton Ochsner, MD, was one of the first to notice the link between smoking and lung cancer. Further studies have estimated that 857c to 907c of lung cancer deaths are caused by smoking. Correa et al5 have doc umented that a greater proportion of residents of southern Louisiana begin smoking at an earlier age, smoke nonfiltered cigarettes, and smoke more per day than north Louisiana residents. This difference between north and south Louisiana can be seen in other lifestyle habits and may be reflected in the north- J south gradient in cancer mortality which is seen in Louisiana.0 45JOURNAL VOL 142 APRIL SO0SS8W BFG07486 Alexendrl* Leleyen* Monro* Shroveport teion Rm|i liki Chiili Nn Orleeni METROPOLITAN STATISTICAL AREA Adjusted to US 1940 age distribution Fig 1. Difference Between US & LA 1980-1984 Av erage Age Adjusted Death Rates Respira tory Cancers -30 ----------------:--------- --------------------------------------------------------------1 AMia*ni liliyam Monro* **" Keep* Late Chart** Now Orlaana METROPOLITAN STATISTICAL AREA Adjusted to US 1940 Age Distribution Fig 2. Difference Between US & LA 1980-1984 Av erage Age Adjusted Death Rates Colorectal Cancers 20855006 fi 1 Deaths attributed to cancer in Louisiana climbed from 850 in 1918 to 8,405 in 1988. While this absolute increase is primarily due to increases in population, it also reflects a trend of increasing rates, particularly in those cancers related to tobacco. Smoking became more common in women in the 1940s, and lung cancer death rates for women began rising in 1964 after the latency period. Historically females in Louisiana, particularly white females, have enjoyed lower risks of cancer than males and also lower rates than females in other parts of the US. Figs 4 and 5 depict female cancer mortality rates by metropolitan statistical area in the state. Only Shreveport, and to a lesser degree New Orleans, have rates above the US average for white females, while only New Orleans and Alexandria have rates above the US average for black females. In Lou isiana, deaths from lung cancer in women have con tinued to rise since 1964, and in 1980 lung cancer killed more white women than breast cancer, the former leading cause of cancer death.7 PANCREATIC CANCER Cancer of the pancreas is still a relatively rare cancer although the mortality rate is growing. Continuing the trend from the 1930s, all sex-race groups except black females have higher pancreas cancer mortality' rates than the US average. More than half of all Lou isiana parishes rank in the top 20 percent of US coun- 46 JOURNAL VOL 142 APRIL Dtith Rates 80co40 - iM 1 20- |h o UE-JEJl lJJinH m - Eaoohaeua Colon Liver Larynx Prostata Melanoma LeuMmii Stomach Rectum Pancreas Lung Bladder 8r*ir> BSfiiHi United States Louiaiana US Cancer Mortality Rates and Tranda Rates per 100.000 Age-adjusted rates per 100,000 US Cancer Mortality Rates and Trends Fig 3. Cancer Death Rates, 1970 to 1979 US and Louisiana White Males ties/parishes with the highest death rates for pan creatic cancer among the whites.4 Fig 6 clearly shows a cluster of pancreatic cancer deaths in the Acadiana area of the state in the 1980s. This same area exhibited high rates in the 1950s and 1960s. An excess risk has been established among cigarette smokers although it is lower than the risks associated with respiratory cancers.6 STOMACH CANCER While mortality from stomach cancer has declined since the 1930s, it continues to be a major cancer prob- BFG07487 -IS --------!------------- .----------------------------------------------------------------------------- 1 AlxasMrla Latym Mor*# Shr*vrf HMD rmi uh Cftariat n Ortaaaa METROPOLITAN STATISTICAL AREA Adjusted to US 1940 Age Distribution Fig 4. Difference Between US & LA 1980-1984 Av erage Age Adjusted Death Rates White Fe males All Cancers METROPOLITAN STATISTICAL AREA Adjusted to US 1940 Age Distribution Fig 5. Difference Between US & LA 1980-1984 Av erage Age Adjusted Death Rates Non White Females All Cancers lem for blacks, especially males, in Louisiana. As with pancreatic cancer, a cluster of high stomach cancer rates can be seen in the Acadiana area. It is postulated that the excessive gastric cancer rate may have re sulted from the interaction of the black and Cajun cultures. Although the most established risk factor is diet, stomach cancer in blacks has also been associated with tobacco.9 BLADDER CANCER While the threat of bladder cancer has been reduced Fig 6. LA Ranked Pancreatic Cancer AADR -- White Males -- 1970-1979 statewide, excessive rates remain in the New Orleans area for all sex-race groups. The rest of the state has rates which are comparable with national rates. LEUKEMIA Excesses in two cancers which were not previously observed in the 1930s have now emerged, leukemia and kidney cancer. All sex-race groups now have ex cessive leukemia mortality. It is difficult to adequately assess the leukemia mortality problem in Louisiana as all types of leukemias are grouped together in na tional mortality reports even though the risks, treat ment, and survival for specific types are vastly dif ferent. It is also important to note that excessive mortality does not necessarily reflect excessive inci dence. Survival is influenced by such factors as rec ognition of symptoms, stage of disease at diagnosis, access to diagnosis and treatment, patient compli ance, and others. The nature of the leukemia problem in southern Louisiana will be more adequately as sessed after incidence rates for the various subtypes of leukemia (as well as all other cancers in southern Louisiana) become available in the spring of 1990 from the Louisiana Tumor Registry. JOURNAL VOL 142 APRIL 47 20855007 BFG07488 KIDNEY CANCER Excessive mortality from kidney cancer is now seen in all sex-race groups except black males. Smoking is associated with kidney cancer as are obesity in women and occupational exposure to substances such as as bestos. Cancer of the renal pelvis has been associated with the long-term use of pain relievers containing phenacetin.8 CONCLUSION Louisiana has very stable patterns in cancer mortality. Those cancers which are responsible for premature mortality in the 1980s were responsible for Louisiana's high cancer rates in the 1930s. More importantly, without intervention, the same cancers will probably continue to be a serious public health problem in 2030. While we cue fortunate in having low' rates for some cancer sites, our respiratory cancer rate is a ser ious burden. It has been well documented that the excess cancer in Louisiana is due in large part to ex posure to tobacco. Innovative methods must be implemented to re duce the number of citizens who are at risk for res piratory cancer. Both prevention efforts for our youth and cessation programs for current smokers are needed if we are to change the cancer mortality trend by 2010. We applaud the Louisiana State Medical So ciety and parish medical societies for their public health education efforts and programs to control smoking. Our efforts must be expanded. REFERENCES 1. Correa P, Chen VW, Craig JF, et al. Cancer in Louisiana. Baton Rouge, La: Louisiana Division of Administration, 1984. 2. Correa P, Chen VW, Craig JF, et al. Cancer in Louisiana, Vol 2. Baton Rouge, La: Louisiana Division of Administration, 1985. 3. Correa P, Fontham , Chen V, et al. Diet, nutrition, and cancer. / La State Med Soc. 1988;1404:43-49. 4. LIS Cancer Mortality Rates and Trends, 3950-2979. Riggan WB, Mason TJ, Van Bruggan J, et al (eds). Washington, DC: National Cancer Institute, Environmental Protection Agency, 1983. 5. Conea P, Johnson WD. Cancer and lifestvle in Louisiana. / La State Med Soc. 1983;135(3):4-6. 6. Con-ea P, Chen VW, Craig JF, et al. Cancer in Louisiana. Vc'. 4. Baton Rouge, La: Louisiana Division of Administration, 1987. 7. Chen VW, Correa P, Zavala D. Lung Cancer: Leading cause of cancer deaths in Louisiana white females. JNCl. 1984;72:1-2. 8. Page HS, Asire AJ. Cancer Rates and Risks. NIH Publication No 85-691, 1985. 9. Fontham ETH, Correa P, Chen V, et al. Tobacco and Cancer 1 La State Med Soc. 1988;1404:29-39. Drs Chen. Correa, and Ms Craig are from the Louisiana Tumor Registry, Office of Public Health at the Dept of Health and Hospitals. Drs Fontham, Chen and Correa are also from the Dept of Pathology at Louisiana State University Medical Center in New Orleans. Mr Ballinger is from the office of Public Health Statistics, Office of Public Health at the Dept of Health and Hospitals. Reprint requests should be sent to Jean F. Craig, Louisiana Tumor Registry. Room 305, PO Box 60630, New Orleans, LA 70360. The JOURNAL also published a cancer issue in April 1988. Included were the following articles: Is Cancer Survival Poorer in Louisiana? Tobacco and Cancer Diet, Nutrition, and Cancer Louisiana Tumor Registry Louisiana Cancer and Lung Trust Fund Board Copies can be obtained from: The JOURNAL of the Louisiana State Medical Society 1700 Josephine Street New Orleans, LA 70113 20855008 48 JOURNAL VOL 142 APRIL BFG07489 EXCESSIVE CANCER RATES AMONG BLACKS IN LOUISIANA: AN OPPORTUNITY FOR PHYSICIAN INTERVENTION VIVIEN W. CHEN, PhD; JEAN FIKE CRAIG, MSHyg; ELIZABETH T.H. FONTHAM, DrPH; PELAYO CORREA, MO 600SS80 18 JOURNAL VOL 142 APRIL BFG07490 Excessive cancer rates among blacks in Louisiana are well-documented. Both male and female blacks have higher overall cancer incidence and mortality rates than their white counterparts. Cancers that are excessive in males include lung, esophagus, larynx, stomach, pancreas, liver, multiple myeloma, and prostate. In black females, higher rates are observed for cancers of the esophagus, stomach, pancreas, liver, multiple myeloma, cervix, and breast (mortality only). The excess of lung cancer among black men is not observed in women. These cancer sites share similar risk factors and are associated mostly with tobacco or diet. Physicians in Louisiana can play an important role in cancer intervention by informing their black patients about the magnitude of the cancer problem in blacks, increased cancer risk associated with tobacco and excessive alcohol use, importance of a balanced nutritious diet, cancer signs and symptoms, and the importance of early detection. lacks in the United States are known to have a B higher overall mortality and to lag behind their white counterparts in life expectancy.1 A sharper in crease in cancer mortality among blacks in the 1950s and early 1960s has placed blacks ahead of whites in deaths from all malignant neoplasms.2 Currently, blacks have the highest overall age-adjusted cancer rates for both incidence and mortality in any US ethnic group.3 Louisiana has a high proportion of blacks in the population, approximately 30% compared to 12% na tionwide.4 Blacks have higher cancer rates than whites, particularly black males. In Louisiana, cancer mor tality in black males varies greatly by geographic area, with a gradient of increase from north to south as demonstrated in the map (Fig l).5 Comparisons of age-adjusted mortality rates for all cancers by Met ropolitan Statistical Area (MSA), along with rates for the state of Louisiana and the United States, 1980 to 1984, are presented in Figs 2 and 3. For nonwhite (over 90% black) males, cancer mortality rates are lower in north and central Louisiana (ie, MSAs of Monroe, Shreveport, Alexandria, and Baton Rouge) than the state and national rates. This is in contrast to south Louisiana, as represented by metropolitan New Or leans and Lake Charles, which has higher rates than both the state and national averages. For nonwhite women, only New Orleans blacks have higher cancer death rates. Fig 1. LA Ranked Cancer AADR--All Sites-- Non-White Males--1970-1979 Aiesendrla Lsttysiis Meeroe Shreveeert ten Roue* Lake Charles New Orleaaa METROPOLITAN STATISTICAL AREA MlsaiaS US (S4S A*# MaiflSaltaa Fig 2. Difference Between US & LA 1980-1984 Average Age Adjusted Death Rates Nonwhite Males--All Cancers. Since the excessive cancer mortality is primarily in New Orleans and sufficient cancer incidence data are currently available only for New Orleans, this pa per will focus on excessive cancer among New Or leans blacks. The cancer profile of New Orleans blacks is compared to New Orleans whites, the blacks in the JOURNAL VOL 142 APRIL 19 o to ssr o e BFG07491 IS METROPOLITAN STATISTICAL AREA U|IIH< I* Ut It4t f> Fig 3. Difference Between US & LA 1980-1984 Average Age Adjusted Death Rates NonWhite Females All Cancers. SEER (Surveillance, Epidemiology and End Results) program of the National Cancer Institute (NCI), blacks in Louisiana and the nation as a whole. It is our hope that a better understanding of the cancer profile in Louisiana will provide a basis for our practicing phy sicians to plan for cancer prevention and cancer con trol. Accordingly, for those cancer sites which have excess incidence, the risk factors are presented since they are pertinent for prevention. For those sites in which the excess rates are restricted to mortality (im plying poorer survival), screening and early detection are emphasized. EXCESSIVE CANCER RATES IN BLACK MALES Table 1 presents cancer sites which are excessive in incidence and/or mortality rates among New Orleans black males as compared to their white counterparts and blacks elsewhere. Lung cancer. Lung cancer is the most common cancer among New Orleans black males. It alone ac counts for more than a quarter (27.5%) of all newly diagnosed cases of cancer and about 30% of all cancer deaths.6-7 The New Orleans black males not only have lung cancer rates (both incidence and mortality) which exceed those of New Orleans white males, but also those of black males in other parts of the United States. They have a higher lung cancer incidence rate than any one registry in the SEER program. In fact, data from Cancer Incidence in Five Continents Volume V 20 JOURNAL VOL 142 APRIL show that New Orleans blacks have the highest rate reported worldwide for 1978 to 1982.8 Cigarette smoking has been well-documented to be the predominant cause of lung cancer, over shadowing all other risk factors. Correa et al sug gested as much as 90% of all lung cancer cases in south Louisiana are attributable to tobacco consumption.*"Low intake of fresh fruits and vegetables was shown to be the second most important risk factor in this population. Other factors predisposing an indi vidual to increased risk for lung cancer include oc cupational exposures to a variety of products such as asbestos, polycyclic hydrocarbons and chromium,10 and also exposure to passive smoking.11 Esophageal cancer. The excess of esophageal cancer risk and mortality among the blacks is very pro nounced. Black males have a threefold excess risk when compared to white males and the mortality rates are twice as high.6-7 Major risk factors for cancer of the esophagus are alcohol intake and tobacco use. Epidemiologic data indicate that the combination of these two factors sub stantially increases the risk of cancer of the esopha gus.12- 13 Other factors associated with increased risk include nutritional deficiencies and possibly con sumption of hot food and drinks which may induce thermal irritation.14 Stomach cancer. Excessive mortality from stomach cancer was observed in both blacks and whites in south Louisiana during the 1950s.15 This excess dis appeared in whites about a decade later but persisted in blacks, resulting in an even larger gap in the racial differential.16 At present, the age-adjusted mortality rate for black males in New Orleans is three times higher than that for white males (19.9/100,000 vs 6.5), exceeding the SEER, state, and US averages. Inci dence data are consistent with mortality, with 2.5 times greater risk among blacks than whites. Low socioeconomic status is correlated strongly with increased rates of stomach cancer. Studies from Louisiana have implicated diet as the major deter minant of stomach cancer risk.17 Consumption of smoked foods, homemade sausages and home cured meats, excessive salt intake, and particularly low con sumption of fresh fruits and vegetables are associated with significant increases in the risk of stomach cancer in blacks. Excess risk is also found among current drinkers and cigarette smokers. Pancreatic cancer. Although black males in New Orleans share similar risk and mortality from pan- BFG07492 8 81 in - K TABLE 1 AGE-ADJUSTED (US 1970) INCIDENCE AND MORTALITY RATES FOR MALES, BY RACE AND GEOGRAPHIC AREA, 1978-81 New Orleans' BM New Orleans WM SEER BM La State BM US BM All sites Esophagus Stomach Liver Pancreas Larynx Lung Prostate Multiple Myeloma 1 M 1 M 1 M 1 M 1 M 1 M 1 M 1 M 1 M 471.4 326.9 18.6 11.6 22.4 19.9 9.7 10.5 15.5 13.5 14.5 4.5 129.7 101.2 99.7 38.5 8.0 4.9 403.5 247.8 5.3 5.5 9.2 6.5 4.6 5.2 9.4 10.8 12.9 3.1 104.7 87.9 58.5 19.7 3.2 2.9 487.9 289.0 19.2 -- 21.3 14.8 5.5 -- 16.8 13.8 12.5 4.7 119.0 93.5 120.3 44.5 10.2 -- I = Incidence rate M = Mortality rate - = NA `Incidence for New Orleans includes Jefferson, Orleans, and St Bernard parishes Mortality for New Orleans includes Jefferson, Orleans, St Bernard, and Plaquemines parishes Sources: Cancer in Louisiana, Vol III, IV, and Cancer Incidence and Mortality in the US 1973-81 274.5 _ 12.1 17.8 _ 6.7 _ 13.8 -- 4.5 -- 88.4 -- 36.0 -- 5.1 286.7 _ 16.0 _ 14.8 _ 5.5 _ 13.5 -- 4.9 -- 91.4 -- 43.9 -- 6.0 creatic cancer as blacks elsewhere in the United States, their incidence rate is 1.5 times higher than the New Orleans whites. The causes of pancreatic cancer are not clear. A study in Louisiana has found excess risk associated with cigarette smoking, consumption of pork products, and low dietary intake of vitamin C.18 Cancer of the larynx. The incidence rate for la ryngeal cancer in black males exceeds the New'"Or leans white rate and the SEER averages by about 15% and has remained stable over the last decade. How ever, the excess in mortality is greater, about 50% higher than New Orleans whites. Tobacco and alcohol consumption are the two major risk factors for laryngeal cancer, and they act synergistically. Other risk factors include occupa tional exposure to asbestos, nickel, and mustard gas.19 Liver cancer. The incidence rate for New Orleans black males rose 47% between 1974 to 1977 and 1978 to 1981 while rates for their white counterparts and rates for black and white males in the nation de clined.6 Currently, incidence and mortality rates are two times higher among blacks than whites. Worldwide, infection with hepatitis B virus is the most important risk factor for liver cancer.20 Alcohol consumption is a major cause for liver cirrhosis which is associated with liver cancer; however, a causal re lationship has not been established.21 Prostatic cancer. Prostatic cancer is the second most common cancer in New Orleans blacks, following lung cancer, with an average age-adjusted incidence rate of 99.7/100,000 and mortality rate of 38.5. National trends in incidence from recent years show that pros tatic cancer has become the leading cancer among men. Despite the fact that black males in New Orleans JOURNAL VOL 142 APRIL 21 N ZTO S S80: BFG07493 TABLE 2 AGE-ADJUSTED (US 1970) INCIDENCE AND MORTALITY RATES FOR FEMALES. BY RACE AND GEOGRAPHIC AREA, 1978-81 New Orleans' BM New Orleans WM SEER BM La State BM US BM All sites Esophagus Stomach Colon Rectum Liver Pancreas Breast Cervix Multiple Myeloma 1 M 1 M 1 M 1 M 1 M 1 M 1 M 1 M 1 M 1 M 287.5 181.6 4.5 3.2 11.8 11.3 33.2 22.0 9.7 5.0 4.3 4.0 9.4 6.6 70.8 29.6 20.0 8.1 4.9 4.6 273.5 138.2 1.6 1.1 4.0 3.8 25.6 16.3 11.5 2.3 3.6 3.2 7.5 7.4 77.3 24.2 8.2 2.8 2.4 1.8 290.3 151.5 5.5 -- 8.3 6.0 36.6 17.7 9.0 2.8 1.8 -- 11.1 9.9 71.9 26.8 20.2 7.3 7.0 -- I - Incidence rate M = Mortality rate - = NA `Incidence for New Orleans includes Jefferson. Orleans, and St Bernard parishes Mortality for New Orleans includes Jefferson, Orleans, St Bernard, and Plaquemines parishes Sources: Cancer in Louisiana, Vol III, IV, and Cancer Incidence and Mortality in the US 1973-81 153.8 -- 2.9 ___ 7.9 ___ 15.7 ___ 2.5 -- 3.5 -- 8.2 24.9 -- 8.6 -- 4.5 152.9 ___ 3.9 ___ 6.6 _ 17.6 ___ 2.8 ___ 2.1 ___ 9.2 ___ 26.3 -- 8.8 -- 4.3 have a lower incidence rate than all registries in the SEER program and a mortality rate 14% below the US average, they still have significantly higher rates than New Orleans whites, with a 70% greater incidence rate, and a 95% greater mortality rate.6'7 The causes of prostatic cancer are not well-estab lished. Studies have suggested increased risk asso ciated with high consumption of dietary fats and pos sible occupational exposure to cadmium.10 Multiple Myeloma. The incidence for multiple myeloma is more than twice as high for blacks as it is for whites, 8.0/100,000 compared to 3.2/100,000. An approximately twofold excess is also observed for mortality. Risk factors associated with the predominance in blacks have not been identified. Higher levels of im munoglobulins in blacks than whites have been sug gested as a possible inborn susceptibility to multiple myeloma.22 Other factors reported to be associated with the disease include some occupational expo sures, ionizing radiation, immunosuppressive dis eases, and more infections in general. EXCESSIVE CANCERS IN BLACK FEMALES Although black females in New Orleans have only slightly higher risk (5% excess) of developing any can cer than white females, the mortality is about 31% higher. Table 2 shows cancers that are excessive in incidence and/or mortality among black females. Breast cancer. Breast cancer is the most common JOURNAL VOL 142 APRIL 23 20855013 BFG07494 cancer for black women in New Orleans. Even though black females have a lower incidence rate, they ex perience a higher mortality rate than white women, a phenomenon not observed nationwide. The ageadjusted mortality rate of 29.6/100,000 is higher than the SEER average, the Louisiana rate, and the national rate, reflecting a poorer survival which is partially explained by more advanced stages at diagnosis.23 In an NCI funded multicenter study of Black/ White Cancer Survival, New Orleans black women were more frequently diagnosed with advanced breast cancer than New Orleans whites. In addition, 45% of New Orleans breast cancer cases were diagnosed with tumor < 5 cm and no lymph node involvement as compared to 53% in Atlanta and 59% in San Francisco. Furthermore, carcinoma in situ accounts for only 4% of the New Orleans cases while it represents 7% and 11% of cases in Atlanta and San Francisco respec tively. These results are consistent with the Louisiana Tumor Registry data that New Orleans women are not at a higher risk of developing breast cancer when compared to the nation but do have a higher rate of dying from breast cancer. Factors associated with early detection and diagnosis include the use of screening mammography, practice of proper breast self-exam ination, recognition of signs and symptoms, and promptness in seeking medical care. The same study showed only 21% of New Orleans breast cancer pa tients (ages 20 to 79) reported having at least one mammogram in the 6-year period prior to diagnosis, compared to 30% in Atlanta and 36% in San Francisco. Cervical cancer. Black females in New Orleans ex perience a 2.5-fold excess incidence and mortality rates from cervical cancer when compared to their white counterparts. Incidence for cervical cancer has been declining in both black and white females in recent years. In Louisiana, the decline has occurred more rapidly for in situ cases than invasive cancer resulting in an in crease in the proportion of more advanced stage. Dur ing the period 1978 to 1981, the proportion of invasive cervical cases was 45% in black women and 36% in white women.6 This pattern contrasts sharply with those observed at other registries, in which a decline in the incidence of invasive cervical cancer is accom panied by a corresponding increase in reported cases of carcinomas in situ, reflecting the benefits of Pap smear screening. The two major risk factors for cervical cancer are 24 JOURNAL VOL 142 APRIL multiple sex partners and early age at first coitus. Recent studies have demonstrated papilloma virus as an etiologic candidate.24-15 Other factors associated with excess risk include lack of cytology screening, cigarette smoking, and a low intake of vitamin C and foladn.10 Cancers of the colon and rectum. In contrast to the cancers of the upper gastrointestinal tract (esophagus and stomach) where a more than twofold excess is observed for blacks, colon cancer exhibits less dis parity between the two racial groups. Black females in New Orleans have a 30% excess incidence rate and a 40% excess mortality rate when compared to New Orleans white females.6 The age-adjusted mortality rate of 22.0/100,000 also exceeds the averages for black females in the SEER program, Louisiana, and the whole United States.7 While the incidence rates of rectal cancer in New Orleans black females are very comparable to the SEER rates and slightly lower than those for New Orleans whites, mortality in females displays a different pat tern. A twofold excess mortality is present in New Orleans black females when compared to any referent group. The excess mortality rates for both colon and rectal cancer implicate a poorer survival among black females in New Orleans. One possible reason for the poor survival is a more advanced stage at diagnosis. The major risk factors for colon and rectal cancer are dietary. Dietary fat intake has been associated with increased risk while consumption of fresh fruits and vegetables and dietary fiber appear to be protec tive.10 Cancers of esophagus, stomach, liver, and pancreas and multiple myeloma. Rates of both incidence and mor tality for these cancers are higher among black females than white females in New Orleans. In addition, New Orleans black females have mortality rates which ex ceed those of the state and United States for cancers of the esophagus, stomach, liver, and multiple mye loma. Since the magnitude of these excesses are sim ilar in males, they are not discussed further. DISCUSSION Excessive cancer rates among blacks in Louisiana are very well-documented. Both black males and females have higher incidence and mortality rates than their white counterparts. A close examination by specific site reveals cancers that are excessive, share similar risk factors and are mostly tobacco26 and diet-related.27 BFG07495 Cigarette smoking is responsible for 30% of all cancers and nearly 90% of lung cancer. It is also a contributory factor for cancer of the oral cavity, lar ynx, esophagus, bladder, pancreas, kidney, and cer vix. Concurrent with a high prevalence of smoking, blacks in Louisiana experience an increased risk for cancers of the lung, esophagus, stomach, pancreas, larynx, and cervix. Louisiana has been known for its distinct ethnic and geographic dietary patterns. These patterns may have an influence on the peculiar cancer profile ob served, especially the high rates of cancers of the res piratory and upper digestive tract. Alcohol consump tion, more prevalent in lower socioeconomic groups, when used together with tobacco increases the risk of laryngeal and esophageal cancer. Consumption of pork products and smoked and home cured meats, which is more prevalent among Louisiana blacks, is associated with an excess risk of stomach and pan creatic cancers in blacks of south Louisiana. Excessive salt intake is suspected as a factor in the high inci dence of stomach cancer in Louisiana blacks. The re cent findings of the protective effect of adequate in take of fresh fruits and vegetables and dietary vitamin C against cancers of lung, stomach, and pancreas is perhaps most interesting and potentially important in cancer prevention. If cancer prevention and control in blacks is to be undertaken in Louisiana, the strongest effort should be directed toward prevention and cessation of to bacco use followed by an emphasis on a balanced diet which includes a higher consumption of fresh fruits and vegetables; less intake of foods that are high in fats, heavily salted or nitrite preserved; and decreased alcohol consumption particularly among smokers. In addition to primary prevention, ie, reduction of risk factors, secondary prevention should also be addressed. This includes early detection for cancers of the breast, cervix, and colorectum by more frequent utilization of screening services such as mammogra phy, Pap smears, and proctoscopic examinations. In a study of attitudes towards cancer and cancer tests sponsored by the American Cancer Society,27 blacks were found to underestimate the cancer prevalence among themselves, to be less likely to recognize can cer warning signs and symptoms, and to participate less in screening programs. They also tended to be more fatalistic and less likely to believe that early de tection is essential and that existing treatment could be effective. Therefore, programs directed at improv ing knowledge of cancer facts, correcting the misin formation on treatment, and promotion of health in terventions are greatly needed. In conclusion, physicians in Louisiana can play an important role in cancer prevention and control among blacks by informing their black patients about: (1) the magnitude of the cancer problem in blacks; (2) the cancer risk associated with tobacco use and ex cessive alcohol consumption; (3) the importance of a balanced nutritious diet which includes more fresh vegetables and fruits, less smoked and salted foods, and less fats; (4) cancer warning signs and symptoms; and (5) the benefits of early detection by screening tests and regular participation in such screening. REFERENCES 1. Health United States 1987. Hvattsville, Maryland: US Dept of Health and Human Services, 1988; PHS Publ No 88-1232. 2. Riggan WB, Van Bruggen J, Mason T, et al. US Cancer Mortality Rata and Trends, 1950-79. Vol I. Washington, DC: NCI/EPA, 1983, 3. National Cancer Institute. 1987 Annual Cancer Statistics Review, including cancer trends: 1950-1985. Washington, DC: US Dept of Health and Human Services, 1988; NIH Publ No 88-2789. 4. Health Status of Minorities and Low Income Croups. US Dept of Health and Human Services, 1985; No (HRSA) HRS-P-DV 85-1, Table 61. 5. Correa P, Chen VW, Craig JF, et al. Cancer in Louisiana. Vol 1. Baton Rouge: Louisiana Division of Administration, Administrative Services, 1983:1-248. 6. Correa P, Chen VW, Craig JF, et al. Cancer in Louisiana, Vol III. Baton Rouge: Louisiana Division of Administrative Service, 1986:1-124. 7. Chen VW, Craig JF, Correa P, et al. Cancer in Louisiana, Vol IV. Baton Rouge, Louisiana: LSU Graphic Service, 1987:1-150. 8. Craig JF, Correa P, Chen VW. USA, Louisiana, New Orleans, 1978-82. In: Muir C, Waterhouse J, Mack T, et al, eds, Cancer Incidence in Five Continents, Vol V. Lyon, France: International Agency for Research on Cancer, 1988:350-357. 9. Correa P, Pickle LW, Fontham E, et al. The causes of lung cancer in Louisiana. In: Mizell M, Correa P, eds. Lung Cancer Causes and Prevention. Deerfield Beach, Florida: Verlag Chemie International, 1984:73-82. 10. Page HS, Asire AJ. Cancer Rata and Risks, 3rd ed. Washington, DC: US Dept of Health and Human Services, 1985; NIH Publication No 85-691. 11. Correa P, Pickle LW, Fontham EH, et al. Passive smoking and lung cancer. Lancet 1983;2:595-597. 12. Tuvns AJ, Pequignot G. Abbatucd JS. Oesophageal cancer and alcohol consumption. Importance of type of beverage. Int J Cancer 1979;23:443447. 13. Pottem LM, Morris LE, Blot WJ, et al. Esophageal cancer among black men in Washington, DC. I. Alcohol, tobacco and other risk factors. JNCl 1981;67:777-783. 14. Pottem LM. Morris LE. Blot WJ, et al. Esophageal cancer among black men in Washington, DC. II. Role of nutrition. JNCl 1981;67:1199-1206. 15. Haenszel W. Variation in incidence and mortality from stomach cancer with particular reference to the United States. JNCl 1958;21:213-262. 16. Mason TJ, McKay FW, Hoover R, et al. Atlas of Cancer Mortality for US Counties: 1950-1969. Washington, DC: Government Printing Office, 1975. 17. Correa P, Fontham E, William PL, et al. Dietary determinants of gastnc cancer in south Louisiana. JNCl 1985;75:645-654. 18. Falk R, William PL, Fontham E, et al. Lifestyle factors for pancreatic cancer in Louisiana. Am j Epidemiol 1988;128:324-336. 19. Austin DF. Larynx. In: Schottenreld D, Fraumeni JF Sr, eds. Cancer Ep idemiology and Prevention. Philadelphia: WB Saunders, 1982:534-563. JOURNAL VOL 142 APRIL 25 20855015 BFG01496 20. Beaslev RP. Hepatitis B virus as the efiolopc agent in hepatocellular carcinoma, epidemiologic considerations. Hepatology 1982;2:218-268. 21. Falk H. Liver. In: Schottenfeld D, Fraumeni JF 5r, eds. Cancer Eptdeimologu and Preivntton. Philadelphia: WB Saunders, 1982:668-682. 22. Blattner WA. Multiple myeloma and macroglobulinemia. In: Schotten* feld D, Fraumeni JF Sr, eds. Cancer Epidemiology and Prei'ention. Phila delphia: WB Saunders, 1982:795-811. 23. Chen VW, Correa P, Craig JF, et al. Is cancer survival poorer in Loui siana? J la State Med Soc 1988;1404:20-28. 24. Zur Hausen H. Condvlomata acuminata and human genital cancer. On cer Res, 1976:36:794. 25. Bosch FX, Munoz N. Human papillomavirus and cervical neoplasia: a critical review of the available epidemiologic evidence. In: Munoz N, Bosch FX, Jensen OM. eds. Human Papillomavirus and Cervical Cancer. Lyon, France: IARC Scientific Pub No 94, 1989:135-151. 26. Fontham ET. Correa P, Chen VW, et al. Tobacco and cancer. J La State Med Soc, 1988;1404:29-40. 27. Correa P, Fontham ET, Chen VW, et al. Diet, nutrition and cancer. J La State Med Soc 1988;1404:43-49. 28. American Cancer Society. Black Americans' attitudes towards cancer and cancer tests: highlights of a study. CA, 1981;31:212-218. Drs Chen, Fontham and Correa are from the Dept of Pathology at Louisiana State University Medical Center in New Orleans. Dr Chen is also from the Dept of Biostatistics and Epidemiology at Tulane University School of Public Health and Tropical Medicine in New Orleans. Ms Craig is from the Louisiana Tumor Registry, OPH, at the Louisiana Dept of Health and Hospitals. Reprint requests to Dr Vivien W. Chen, Dept of Pathology, LSU Medical Center, 1901 Perdido St, Neu.' Orleans, LA 70112. Work supported by Grant #N01-CP-43262 and the Louisiana Dept of Health and Hospitals. OBSTETRICS & GYNECOLOGY Excellent opportunity for a board-certified OB/GYN. Teaching position available in South Louisiana, 55 miles from New Orleans. Program con sists of 16 residents, 6 of whom are involved at all times in Ochsner Medical Foundation integrated in stitution located in New Orleans. Very active prac tice composed of high-risk obstetrical and gynecological surgical service patients. Present staff consists of one fulltime board certified OB/GYN with excellent consultative ability and major subspecialties. Very competitve salary with superb fringe benefits. Contact: Dr. Alvin H. Lassen Medical Director South Louisiana Medical Center 1978 Industrial Blvd Houma, LA 70363 (504)873-1265 j ^ ! 9Tnccoo> Studies on cancer are highlighted this month in honor of Cancer Month. Coor dination of this issue was provided by the Louisiana Cancer and Lung Trust Fund Board. The Board was established in 1980 to oversee the Louisiana Tumor Registry and to encourage, through the awarding of grants, research on the cancer pro blem in Louisiana. Three of the grants funded by the Board are reported on in this issue: "The Louisiana Cancer Consortium"; "Tobacco, Alcohol, and Marijuana Use Among Black Adolescents"; and "Excessive Cancer Rates Among Blacks." 26 JOURNAL VOL 142 APRIL BFG07497 PEDIATRIC CANCER IN NEW ORLEANS FRANK D. GROVES, MD; JEAN FIKE CRAIG, MSHyg; VIVIEN W. CHEN, PhD; ELIZABETH T.H. FONTHAM, DrPH; DIEGO E. ZAVALA, MSc; PELAYO CORREA, MD Rates for pediatric cancer in the Greater New Orleans area were compared with rates from the National Cancer Institute's SEER (Surveillance, Epidemiology, and End Results) Program. The same patterns observed nationally were seen in New Orleans over a 10-year period. Rates were comparable with the exception of cancers of the brain and central nervous system for which New Orleans children displayed higher rates. Using the large number of cases in the SEER Program, three etiological patterns of childhood cancer were apparent based on the age at diagnosis. Cancer in children, though low in incidence com pared to adult cancers, is second only to acci dents as the leading cause of death among children less than 15 years old. Childhood cancer incidence rates for the metro politan New Orleans area have been determined by the Louisiana Tumor Registry and compared with na tional rates calculated by the SEER (Surveillance, Ep idemiology, and End Results) Program of the National Cancer Institute. Because childhood cancer is rela tively rare, 10 years of data in New Orleans (1974 to 1983) were used to ensure stable rates. The incidence rates are reported as per million child-years. Incidence rates were calculated by age group (0 to 4, 5 to 9, and 10 to 14), as well as by race-sex groups so that dif ferences between age groups can be examined. The Manchester Classification System used in this manuscript classifies pediatric tumors according to histology, rather than anatomic site. This system is utilized because most childhood cancers are leuke mias, lymphomas, sarcomas, or primitive embryonal cell lines which are not necessarily limited to a specific part of the bodv as are the carcinomas in adults. JOURNAL VOL H2 APRIL 27 0855017 BFG07498 TABLE AVERAGE ANNUAL AGE-ADJUSTED (US 1970 STANDARD) PEDIATRIC CANCER INCIDENCE RATES PER MILLION CHILD-YEARS BY RACE/SEX GROUPS IN NEW ORLEANS AND THE SEER PROGRAMS 1974-1983 Leukemia Lymphoma Brain & CNS Neuroblastoma Retinoblastoma Renal Tumors Hepatic Tumors Malig. Bone Tumors Soft Tissue Germ-Cell, etc. New Orleans SEER WM BM WF BFs WM BM WF BF 39.0 39.9 21.0 12.1 43.0 23.4 35.9 24.4 20.1 19.1 11.2 3.5 21.8 14.9 11.6 6.9 29.6 26.0 26.5 29.4 24.0 18.5 20.6 19.8 10.7 5.2 7.0 8.6 9.4 8.4 9.6 8.9 5.4 3.5 1.4 5.2 3.1 3.9 3.6 4.3 9.4 8.7 5.6 8.6 7.1 10.4 8.3 12.5 1.3 0.0 0.0 0.0 1.7 1.3 1.2 1.3 5.4 10.4 4.2 3.5 6.0 4.9 5.7 5.6 10.7 6.9 7.0 5.2 9.5 9.1 8.3 8.9 2.7 0.0 1.4 5.2 3.5 2.3 4.4 5.3 INCIDENCE OF PEDIATRIC CANCERS IN NEW ORLEANS Although a few pediatric tumors are more common in a certain race or sex group (such as Ewing sarcoma in white children), the patterns observed in New Or leans children are the same as patterns seen nation ally. Leukemias are the most frequent childhood ma lignant neoplasms, followed by brain tumors, lymphomas, soft tissue sarcomas, neuroblastomas, bone tumors, and Wilms tumors. Leukemia. More than 75% of the pediatric leuke mias are acute lymphoid leukemias both locally and nationally. Leukemias show a gender variation, with boys having a two-to-threefold excess over girls (Ta ble). Of the four race-sex groups, only black boys in New Orleans have rates higher than the national av erage but they are not statistically significant. Among girls, local rates are significantly lower, by almost 50%, than the SEER rates. Lymphoma. Similar to leukemias, lymphomas also show a male excess in both New Orleans and the SEER program. Other national patterns for lympho mas are also observed in New Orleans with compa rable rates except for a deficit in black girls. The pe diatric lymphomas are evenly divided among three histologic categories: Hodgkin, nonHodgkin, and other, which includes Burkitt lymphoma, histiocy tosis, and other. Brain and Central Nervous System. Children of all four sex-race groups display higher brain and central nervous system cancer rates locally than are seen na 28 JOURNAL VOL 142 APRIL tionally although the excess is not statistically signif icant. No race or sex predilection is apparent. Bone. New Orleans black boys have a twofold excess of bone cancer compared to black boys in the SEER program, but the difference is not statistically significant. Rates for the other demographic subgroups approximate the national averages. Among white children of both sexes, the bone tumors are evenly divided between Ewing sarcomas and osteosarcomas. Among blacks, however, Ewing sarcomas are rare, and the vast majority of bone tumors are osteosar comas. Other Pediatric Cancers. Small numbers of cases for other pediatric cancers such as hepatoma, neu roblastoma, retinobalastoma, Wilms tumor, soft tis sue sarcoma, and germ cell cancer result in unstable rates which appear comparable to national rates. In fact, only one childhood liver case was reported in the last 10 years. PATTERNS OF AGE AT DIAGNOSIS Since the age at diagnosis can provide clues to the timing of exposure to carcinogenic insults, the pat terns of age at diagnosis were examined. Because of the small numbers of cases in New Orleans, partic ularly the rare cancers, data from the SEER program were used. Three distinct patterns related to age at diagnosis can be observed. In Fig 1, the prenatal in fluence can be seen in pediatric neuroblastoma with the greatest number of cases occurring in the first year of life. Retinoblastoma, which has been shown to have 20855018 BFGO7499 40 10 Fig 1. PEDIATRIC NEUROBLASTOMA, 1974-1983. Seer Program, All Races, Both Sexes. Distribution by Age at Diagnosis. o 1 2 3 4 S 6 7 8 9 10 11 12 13 14 Years of Aae Fig 3. PEDIATRIC BRAIN CANCER, 1974-1983. Seer Program, All Races, Both Sexes. Distribution by Age at Diagnosis. Years of Age Fig 2. PEDIATRIC BONE CANCER, 1974-1983. Seer Program, All Races, Both Sexes. Distribution by Age at Diagnosis. a genetic etiology, also fits this pattern as well as Wilms tumors and acute leukemias although their de cline is more gradual and later than the second year of age. The second predominant pattern shown in Fig 2 is characterized by low postnatal frequencies and gradual elevations in the preschool and school-age years, reaching a peak by the preadolescent years. This pattern is most typically manifested by malignant bone tumors. It is also observed for malignant lym phomas, in marked contrast with leukemias which also originate in the hematopoietic tissue. This pattern suggests postnatal carcinogenic influences and is probably also related, in the case of bone tumors, to the fact that the target tissues are in a state of active proliferation and therefore are most susceptible to car cinogens. A peculiar combination of the postnatal and preadolescent peaks is observed for gonadal tumors, especially the germ-cell neoplasms. In boys, such tu mors are predominantly postnatal whereas in girls they are predominantly preadolescent with a smaller peak. The predominant tumor in boys is the infantile embryonal carcinoma (also called "yolk-sac tumor" or endodermal sinus tumor), whereas in girls the pre dominant tumor is the dysgerminoma which resem bles the male seminoma. This observation suggests that the target cell, rather than the timing of the car cinogenic insult, determines the epidemiologic pat tern in the case of gonadal tumors. The absence of postnatal and preadolescent peaks constitutes a third (flat) pattern of pediatric cancer as seen in Fig 3. This pattern is observed in tumors of the central nervous system and in soft-tissue sarco mas. Both tumors represent mostly support tissues, and the flatness of the frequencies suggests random "hits" of carcinogenic forces on nonspecialized tissues at a rather constant rate of replication. Some excep tions to the above rule are the rhabdomyosarcomas and the ependymomas which represent more spe cialized tissues and have a postnatal peak. 29JOURNAL VOL 142 APRIL BFG07500 6TOSS80Z