CancerIncidenceand Survivalin Patients65 Yearsof Age and Older

Transcription

1 CancerIncidenceand Survivalin Patients65 Yearsof Age and Older Anne BaranoVsky, MS. Max H. Myers, Ph.D. Introduction Cancer statistics vary greatly by age as well as by type of cancer. These age groups are usually categorized as: children (below the age of 15), young adults (15 to 44), middle age (45 to 64), and the elderly (65 and older). Although considerable research has been devoted to cancers of childhood, there have been few studies of cancers in the elderly, despite the fact that over 50 per cent of all cancers occur in this age group. This paper presents data from the Sur veillance, Epidemiology and End Results (SEER) program, concerning cancer inci dence and survival in persons age 65 and older. To gain a perspective on the chang ing impact of cancer with increasing age, we will compare these statistics with those for persons age 45 to 64. Material and Methods This report is based on data for all primary invasive cancers, excluding basal or squa Ms. Baranovsky is Statistician in the Operations Research Section (Surveillance and Operations Research Branch) of the National Cancer Insti tute in Bethesda, Maryland. Dr. Myers is Statistical Consultant to the Biom etry Branch of the National Cancer Institute in Bethesda, Maryland. The authors acknowledge the contributions of the SEER staff of the National Cancer Institute and the SEER registries for collection and prep aration of the data, mous cell skin cancers, diagnosed among persons 65 years of age or older in eight geographic areas participating in the SEER program from 1973 through These population-based reporting areas include the states of Connecticut, Iowa, New Mex ico, and Utah, and the metropolitan areas of Detroit, Atlanta (since 1975), Seattle (since 1974), and San Francisco-Oakland.' For computing the incidence rates, es timates of the SEER white population are derived from the actual census counts by the US Bureau of the Census in 1970 and 1980 for all registries except New Mexico. Estimates for New Mexico came from the New Mexico Tumor Registry. Average an nual incidence rates are based on cases di agnosed during the periods 1973 to 1975 and 1979 to All incidence rates are age-adjusted by the direct method, using as the standard the 1970 US population age 65 and older. Survival results, based on first primary cancers diagnosed from 1973 through 1979, are presented as five-year observed and rel ative survival rates, specific for sex and cancer site.2-3the relative rate is the ratio of the observed survival rate for cancer patients to the expected survival rate for a population similar to the cancer patients with respect to age, sex, race, and year of observation. The relative rate is an esti mate of the chances of surviving the effects of a particular cancer; survival is adjusted, therefore, to take into account deaths due to causes other than cancer. 26 CA-A CANCER JOURNAL FOR CLINICIANS

2 Incidence rates take into account all primary invasive cancers, including those diagnosed solely by death certificate or au topsy. Survival rates, however, exclude cases diagnosed by death certificate or au topsy, and include only cases of first pri mary cancers under active follow-up. Ninety-three percent of cancers were mi croscopically confirmed, ranging from 96 percent for patients 65 to 74 years of age to 81 percent for those 85 and older. Fewer than three percent of cancers were based solely on autopsy or death-certificate di agnosis. Eighty-nine percent of the inci dent cases were first primary cancers. Data are initially presented for 38 spe cific forms of cancer as classified by the International Classification of Diseases for Oncology (ICD-O).4 Since it is not feasible to present a detailed analysis for every can cer site, other analyses are limited to either 10 or 15 of the more common cancer sites. For the 10 most frequent cancer sites during 1979 to 1981, the differences in incidence by sex and by age (65 to 74 versus 75 and older) are examined. These sites are lung, colon, prostate, female breast, urinary bladder, rectum, uterine corpus, pancreas, stomach, and leukemia. This report also presents trends in av erage annual incidence rates between 1973 to 1975 and 1979 to 1981 and five-year survival rates, both observed and relative, for those with cancer diagnosed between 1973 and 1979, by age (65 to 74, 75 and older, and the total 65 and older) and sex. The incidence trends are based on data from the six SEER registries that contributed data for both entire time periods (i.e., data for Atlanta and Seattle were excluded). For these analyses, five more cancer sites were added to the initial 10 sites: larynx, uterine cervix, ovary, kidney, and non-hodgkin's lymphomas. Leukemia was subdivided into acute and chronic lymphocytic, acute and chronic granulocytic, and â œ other leukemiaâ for the survival data and the overall descriptive data on incidence. The last category in cludes lymphocytic or granulocytic leu kemia not specified as acute or chronic, monocytic leukemia, acute or chronic leu kemia with type not specified, aleukemic or subleukemic leukemia, and leukemia not otherwise specified. The category â œ uterinecorpusâ in cludes cases classified as uterus not other wise specified,5-6 which accounts for only three percent of the cancers included in this group. The kidney site also includes can cers of the ureter, urethra, and urinary sys tem not otherwise specified. The other and unspecified sites are almost entirely un specified cancer sites (99 percent), but also include some miscellaneous reticuloen dothelial and myeloproliferative and lym phoproliferative diseases. Results Table 1 shows the extent to which cancer incidence affects the population 65 and older. From 1979 to 1981, the average an nual age-adjusted incidence rates per 100,000 were 2,468.2 for males and 1,401.1 for females. The age group 65 and over accounted for 59 percent of all can cers diagnosed among males and 52 per cent of all cancers diagnosed among females. The five leading cancer sites for white males age 65 and older were prostate, lung, colon, urinary bladder, and rectum. About For elderly men, the incidence of all cancers combined is four times that for men between the ages of 45 and percent of all cancers of the prostate and 55 percent of lung cancers were di agnosed among males in this age group. For colon, urinary bladder, and rectum cancers, 68 percent, 64 percent, and 62 percent of cases, respectively, were diag nosed among males 65 and older. The five leading cancer sites for fe males, in descending order, were breast, colon, lung, uterine corpus, and rectum. The age group 65 and over accounted for 50 percent of all cases of cancer of the lung and somewhat less than 50 percent of VOL 36, NO 1 JANUARY/FEBRUARY

3 â œ J â 14 -c -.,. N.. U

4 CO U) N (C N 0 C') CO (0 C') N (0 N N N U) U) U) N C') N P CO (0N C') â -N4-N NN â -N COCO COâ - COâ - NN U)(0 CO NC') â U) â -(0(0 U) (0â -â -@.U) C)(0 C) 0â - N0 Nâ - CO a) NU) NC') N0 N U) C')C')C') U)(0 (0 â â â U) U)U) (0C') -U)(0â -U)(0C')a)a)@oqU) (0(0 NC') (0N (0COa)U)(0 U)(0 U)â - N(0 C')(0 Na) NC')â - C')N â -(0â (0COâ -Na)NCO(0 0 COâ - C) C')N U) â CO U) NN N U)N a) â -(0 0 â -C') U) a)0 N 00 N NC') U)N 0 NCO â - (0U) CO.e e@ Cd 5, -@..@ C --.C CD U..@.t ee a - â œ a 00 >.C >@ CC 5,.5 -@ -@ -@.@ 2.E CD 5) CD C 0 CD 4-' o@ 0 CO 0 0 VOL 36.NO 1 JANUARY/FEBRUARY

5 @ 0 â c â â c-@ â II Ec@ zo 85< â a â c -@ D â

6 CO 0 CO N N N N 0 U) â - a) C') â - N N U) U) (0 N U) N U) U) (0 N(0 (0N.-0 COCO 0â -C COU) CO C')CO a)n, a).n C')N 0C') N â -U) CONP COC') (0 COC') C)a) N NU) NN NU) â -CO â N â -a) (0 C') â a) C')0 a) (0CO N N â -(0 NCO U)N â C')CO N â -0 C')C') N C')a) NN NN N(0 â (0 (0CO (0NCOâ -U)(0@â NC')@-0(0U)(0â -C')N N (0C') U)C') â -(0N U)U)CO U).- COC') NU)(0â - (0C)N â -(0 N0) (0@-N(0 â -CON â -(0 N0 C') U)N.-.-C')a) â -U) a) C')(0. â -a) N NC) U)U) U)â -(0 U) U)â - N C')â a) â -U) a) C')â U) C) >â U. CD U. U. E @.-.C -@ >.@ > -@ D. E 5) CD 0.C 0 â -â C U. E.C a C D C U. U. â CD 0 â CD - a E C CD â a U. 00 C 0 -â â â E â CD C) C C @ W CO Iâ 0 I 2 I- 0 â 4- VOL 36,NO 1 JANUARY/FEBRUARY

7 cancers of the breast and uterine corpus. About three fourths of all female patients with colon cancer and two thirds of those with rectal cancer were 65 and older. Male incidence rates were higher than female rates for every cancer site common to both sexes, except gallbladder, thyroid gland, and breast. Male rates were about four times higher than female rates for can cers of the urinary bladder and lung, and seven times higher for cancer of the larynx. Males had about double the risk of females for cancers of the stomach and kidney and for leukemia, and a little less than double the risk for cancer of the rectum. Leukemia incidence rates were about twice as high for males as for females in each of the subcategories of leukemia, and males also had higher rates for non Hodgkin's lymphomas than did females. Male incidence rates were 2.5 times greater than female rates for cancer of the liver, while female rates were almost twice those for males for cancer of the gallbladder. The figure shows the difference in cancer incidence by age for the 10 leading sites of cancer. For males, the incidence rates for ages 75 and over were, in general, 1.5 to two times those for the 65-to-74 age group. A notable exception was lung can For elderly women, the incidence of all cancers combined is twice that for women between the ages of 45 and 64. cer, for which the ratio was For fe males, the differences in incidence by age were similar to those for males, with the exception of breast cancer, for which the age-incidence ratio was only 1.11; lung cancer, for which the ratio was 0.73; and uterine corpus cancer, for which the ratio was Trends over time in cancer incidence are indicated by comparing the average an nual incidence rates for 1973 to 1975 with those for the latest years, 1979 to 1981 (Table 2). This comparison is based on data from the SEER areas that had data available for both time periods. Even in this rather brief time span, there are indi cations of change. For males, there were increases of about two percent per year or greater for cancers of the colon, lung, pros tate, and kidney, and for non-hodgkin's lymphomas; there were decreases of about two percent per year for cancer of the pan creas and for leukemia. For females, the incidence rates in creased substantially for cancer of the lung and for non-hodgkin's lymphomas, with annual increases of 10.1 and 4.1 percent, respectively. (The seemingly larger annual increase for laryngeal cancer for females was due to the very low incidence rates). Of particular interest are decreases in in cidence for cancer of the uterine cervix (4.5 percent per year) and leukemia (2.5 percent per year). For many of the sites common to both sexes, there was a con sistent change over time in incidence. For lung cancer and non-hodgkin's lympho mas, however, the incidence rates for fe males increased more rapidly than those for males. The decrease in incidence for leukemia among females was more rapid than that observed for males. Another measure of the impact of can cer on an elderly population is the length of survival after diagnosis of cancer. Table 3 shows the five-year survival rates for each of 15 selected sites, with leukemia shown in detail by type. The observed sur vival rates were usually much lower than the relative survival rates, because patients in these age groups have high risks of dying of diseases other than cancer. For all can cer sites combined, the five-year relative survival rate for males 65 and older was 38 percent, compared with 45 percent for females. This overall female-to-male dif ference was due in part to differences in distribution in site, since for many sites, females had only a slight survival advan tage. Of interest are lung cancers and the chronic leukemias, both lymphocytic and granulocytic, for which the female relative survival rates were higher than those for males, and cancers of the bladder and kid ney, for which the female rates were lower than those for males. 32 CA-A CANCER JOURNAL FOR CLINICIANS

8 Male 10,000 All Sites 0 0 d 1,000 0 a Prostate Lung Colon 5, Bladder Rectum 5, 0 Stomach C 5, 100 Leukemia 0 C Pancreas 10 I I Age at Diagnosis Female 10,000 0 All Sites 8, 0 1,000 Breast Colon Lung U C Rectum 5, 100 â.9 Corpus U C Pancreas Bladder Stomach Leukemia Age at Diagnosis Average annual age-adjusted incidence rates per 100,000, by cancer site and age. for white patients 65 and older, SEER, 1979 to VOL 36. NO 1 JANUARY/FEBRUARY

9 Primary Site All sites 1.3â il.5 1, CA-A CANCERJOURNALFORCLINICIANS

10 Number Relative Observed Number Relative Observed Primary Site and Age of Rate Rate of Cases Rate Rate VOL 36, NO 1 JANUARY/FEBRUARY

11 Number Relative ObservedNumber Relative Observed Primary Site and Age of Cases Rate Rate of Cases Rate Rate - 36 CA-A CANCER JOURNAL FOR CLINICIANS

12 Number Relative Observed Number Relative and Age of Cases Rate Rate of Cases Rate Rate VOL 36, NO 1 JANUARY/FEBRUARY

13 Primary Site and t i@ber Relative Observed Number Relative Observed Rate Rate of Cases Rate Rate 38 CA-A CANCER JOURNAL FOR CLINICIANS

14 Primary Site Inci dence Ratio 65 and 65 and Olden 45 to 64 Older 45 to to and Older Inci dence Ratio 65 and Olden 45 to 64 All sites , , VOL 36,NO 1 JANUARY/FEBRUARY

15 Primary site 65 and 45 to 64 Older All sites 37'_ I' 14, J -â â -49 â L@f'.â 'iâ : œ: i._. IJ tt ii - F., - -@ - â Kidrià y@ lqkin's â œ mphomas 1_I-i- - I -@ 40 CA-A CANCER JOURNAL FOR CLINICIANS

16 Relative survival, which adjusts for deaths due to causes other than cancer, was lower at ages 75 and over than at ages 65 to 74 for cancers of the rectum, larynx (males only), uterine cervix, uterine cor pus, prostate, urinary bladder, and kidney (females only), and for non-hodgkin's lymphomas and the chronic leukemias, both lymphocytic and granulocytic (females only). For many cancer sites, however, the age of elderly patients at diagnosis did not seem to affect relative survival. Survival rates for patients 75 and older were little different from those for patients 65 to 74 years of age for cancers of the stomach, colon, pancreas, lung, female breast, ovary, and kidney (females only), and for the acute lymphocytic and acute granulocytic (males only) leukemias. Summary and Conclusions The impact of cancer on persons 65 years of age and older has been assessed by ex amining incidence rates and survival rates. For all cancers combined, the incidence rate shown in Table 4 for males 65 and older (2,468.2 per 100,000) is four times the age-adjusted rate for males 45 to 64 years of age (586.7). For elderly females, the incidence rate is twice that for females aged 45 to 64 (1,401.1 versus 609.7). Ra tios of incidence rates for older versus younger males are about four to five for cancers of the stomach, colon, rectum, pancreas, and urinary bladder, and for leu kemia; about three for cancers of the lung and kidney, and for non-hodgkin's lym phomas; and 10 for cancer of the prostate. For females, the corresponding ratios are similar to those for males, although a little lower for cancers of the colon, rectum, and urinary bladder, and for leukemia, and a little higher for cancers of the stomach and pancreas. The ratios for breast, uterine cer vix, uterine corpus, ovary, and lung are less than two. The relative survival rates for patients 65 and older are for many cancer sites only a few percentage points lower than rates for those 45 to 64 years of age (Table 5), suggesting that patients in this age group fare only a little worse than younger pa tients in escaping the effects of cancer once it has been diagnosed. Exceptions are cancer of the urinary bladder and non Hodgkin's lymphomas for both men and women and cancers of the uterine cervix, uterine corpus, ovary, and kidney for women. For these sites, the survival rates for older patients are considerably lower than for their younger counterparts. For female breast cancer patients, there was no difference in the five-year relative survival rate for those 65 and older compared with those 45 to 64. References 1. Horrn JW, Asire AJ, Young JL Jr, et al (eds): SEER Program: Cancer Incidence and Mortality in the United States 1973â 81, NIH Publication No. 85â 1837.Bethesda, Md, National Cancer Institute, Cutler SJ, Ederer F: Maximum utilization of the life table method in analyzing survival. J Chronic Dis 8:699â 712, Ederer F, Axtell LM, Cutler SJ: The relative survival rate: A statistical methodology. Nail Cancer Inst Monogr 6:101â 121, ICD-O. International Classification of Dis eases for Oncology. Geneva, Switzerland, World Health Organization, Bailar JC Ill, Eisenberg H: Uterine tumors of unspecified origin. Cancer 18:589â 591, Marrett LD, Meigs JW, Flannery it: Trends in the incidence of cancer of the corpus uteri in Connecticut, 1964â 1979,in relation to con sumption of exogenous estrogens. Am J Epi demiol 116:57â 67,1982. VOL 36, NO 1 JANUARY/FEBRUARY