Is a liquid-based cytology more sensitive than a conventional Pap smear?

BACKGROUND: The comparative sensitivity of liquid-based cytology (LBC) test and conventional Papanicolaou (Pap) smears is controversial. MATERIAL AND METHODS: This study analyses the distribution of cytology, histology, colposcopy and large loop excision of the transformation zone among women screened in Iceland with LBC at the Cancer Detection Clinic in Reykjavik and with a conventional Pap smear outside the Detection Clinic in 2007-2011. The study material included 42 654 LBC tests from 20 439 women and 103 909 Pap smears from 61 574 women. The period 2000-2004 is used to correct for potential bias as a result of unequal distribution of the studied parameters between the study sites before the introduction of LBC. RESULTS: The observed results indicated that women screened with an LBC sample had significantly decreased detection rates of inadequate smears, increased detection of low-grade squamous intraepithelial lesion (LSIL)/atypical cytology and referrals to colposcopy, and an increased detection rate of cervical intraepithelial neoplasia grade 2 or worse (CIN2+) irrespective of age. LBC increased significantly the detection rates of high-grade squamous intraepithelial lesion or worse (HSIL+) cytology and CIN3+ histology only in women under 40 years of age. Taking into consideration the unequal prevalence of the studied parameters between the study sites in 2000-2004 indicated, however, that LBC only affected the rate of inadequate and low-grade cytology tests under the age of 40 years. Positive predictive values for CIN2+ were not significantly different between the tests. CONCLUSIONS: The study results support the view that LBC is no more sensitive than Pap smears for the detection of HSIL+ and CIN2+ irrespective of age. LBC decreased the rate of inadequate smears, but increased the rate of low-grade cytology under the age of 40 years and decreased the total rate of abnormal smears over the age of 40 years.

Cervical cancer: cytological cervical screening in Iceland and implications of HPV vaccines.

This paper reviews the Icelandic experience regarding the age-specific effectiveness, optimal targeted age range and intervals in cervical cancer screening and the screening implications of the HPV16/18 vaccines. The background material is based on data from a screening programme with centralized records dating back to 1964, as well as from population-based studies on the distribution of oncogenic HPV types in cancer and histologically verified CIN2-3 lesions and from the Icelandic arm of the Future II trial with Gardasil. The findings confirm significant increased rates in the screened population of CIN2-3, stage IA (microinvasive) cancer since 1979, mainly in the age group 20-34 years. These lesions start to accumulate within 3 years of a normal smear. Studies on the distribution of HPV types indicate that the marketed vaccines could lower the incidence of cancer and CIN2-3 by about 67% and 53%, respectively, after taking into account reported cross-protection. About 65% of women below 25 years of age had lesions related to the non-vaccine types and after the last normal smear these cases accumulated at the same frequency as cases with vaccine-included types. Cases with combined vaccine and non-vaccine types accumulated at a slower rate. We conclude that screening should continue to start at age 20 years, with invitations at 2-year intervals up to age 39 years and thereafter at 4-year intervals up to age 65-69 years. Current data support the conclusion that the optimal age for catch-up HPV vaccination should be considered in the context of sexual practices and the data do not support changes in the lower age limit or screening intervals for the vaccinated women.

The impact of a quadrivalent human papillomavirus (types 6, 11, 16, 18) virus-like particle vaccine in European women aged 16 to 24.

BACKGROUND: Quadrivalent human papillomavirus (HPV types 6/11/16/18) L1 VLP vaccine is highly effective in preventing HPV 6/11/16/18-related cervical and external genital disease. Herein, we evaluated the impact of the quadrivalent HPV 6/11/16/18 L1 VLP vaccine on prevention of HPV-associated cervico-genital lesions in a broad population of sexually active European women. METHODS: Female subjects (N = 9265) aged 16-24 with four or fewer lifetime sexual partners were enrolled and randomized to quadrivalent HPV vaccine or placebo. Subjects underwent cervicovaginal sampling for HPV DNA detection. Papanicolaou testing and anti-HPV 6/11/16/18 serology testing was also performed. RESULTS: Vaccine efficacy against lesions representing immediate cervical cancer precursors (cervical intraepithelial neoplasia grade 2/3 or adenocarcinoma in situ) related to HPV 6/11/16/18 in the per-protocol population was 100.0%[95% confidence interval (95% CI), 89.8-100.0]. Efficacy against external genital lesions (vulvar or vaginal intraepithelial neoplasia, condyloma, vulvar or vaginal cancer) related to vaccine HPV types in the per-protocol European population was 99.0% (95% CI, 94.4-100.0). CONCLUSION: These data demonstrate that quadrivalent HPV 6/11/16/18 vaccination programs in 16- to 24-year-old European women can be beneficial. NCT0009252, NCT00092534, NCT00092495.

A case-control study to estimate the impact of the Icelandic population-based mammography screening program on breast cancer death.

BACKGROUND: The Icelandic breast cancer screening program, initiated November 1987 in Reykjavik and covering the whole country from December 1989, comprises biennial invitation to mammography for women aged 40-69 years old. PURPOSE: To estimate the impact of mammography service screening in Iceland on deaths from breast cancer. MATERIAL AND METHODS: Cases were deaths from breast cancer from 1990 onwards in women aged 40 and over at diagnosis, during the period November 1987 to December 31, 2002. Age- and screening-area-matched, population-based controls were women who had also been invited to screening but were alive at the time their case died. RESULTS: Using conditional logistic regression on the data from 226 cases and 902 controls, the odds ratio for the risk of death from breast cancer in those attending at least one screen compared to those never screened was 0.59 (95% CI 0.41-0.84). After adjustment for healthy-volunteer bias and screening-opportunity bias, the odds ratio was 0.65 (95% CI 0.39-1.09). CONCLUSION: These results indicate a 35-40% reduction in breast cancer deaths by attending the Icelandic breast cancer screening program. These results are consistent with the overall evidence from other observational evaluations of mammography-based programs.

Risk variables affecting high-grade Pap smears at second visit: effects of screening interval, year, age and low-grade smears.

Our aim was to study the effect, during the period 1979-1996, of the potential risk factors (i) year, age at second visit and first screening interval on the frequency of detection of low- and high-grade smears at the second visit after a normal smear at the first visit; (ii) year, age at second visit and low-grade smears at first visit on the detection of high-grade smears at second visit; (iii) detection of low- and high-grade smears by calendar year at second visit after a normal first visit (period 1981-1996); (iv) proportion of high-grade smears at second visit attributable to low-grade smears at first visit (exposed group); and (v) effect of increasing the screening interval from 2 to 5 years. The results were as follows: (i) low-grade smears increased significantly with years, high-grade smears increased significantly with screening interval and both grades decreased significantly with age; (ii) high-grade smears increased significantly with low-grade smears at first visit and with year but decreased significantly with age; (iii) a significant increase in low-grade smears at second visit with years; (iv) 97% of high-grade smears at second visit were attributable to low-grade smears at first visit; and (v) the risk of high-grade smears was 60% higher when the screening interval was 5 years rather than 2 years after a normal visit at age 20. The strongest risk factors for high-grade cell changes were low-grade smears at first visit [odds ratio 10.2 (ii) and 29.0 (iv)] and first screening interval [odds ratio 1.6 (v)].

The Icelandic and Nordic cervical screening programs: trends in incidence and mortality rates through 1995.

BACKGROUND: The objective of cervical cancer screening is to lower the incidence and mortality rates of the disease. This study evaluates the effectiveness of cervical screening and the UICC and EC screening recommendations based on the Nordic screening experience. METHODS: The study analyzes the features of the Icelandic and the Nordic screening programs and the observed trends in the incidence and mortality rates in these countries through 1995. RESULTS: Organized screening started in all the Nordic countries soon after 1960 and had nation-wide coverage in all these countries, except in Denmark (45% coverage in 1991), by around 1973 but in Norway screening was only spontaneous up to late in 1994. Up to 1985 the target age group and screening interval were most intensive in Iceland, followed by Finland, Sweden and Denmark. All countries except Finland lowered the lower age limit and intensified the screening intervals after 1985. Through the period 1986-1995 the reduction in both the mortality and the incidence rates was greatest in Iceland (mortality: 76% and incidence: 67%) and Finland (73% and 75%, respectively), intermediate in Sweden (60% and 55%, respectively) and Denmark (55% and 54%, respectively), and lowest in Norway (43% and 34%, respectively). The age-specific incidence in the 20-29 age group has been increasing since 1971 in all the Nordic countries, except in Finland, where the yearly registered age-specific incidence has been increasing in the targeted 30-54 age group since 1991. In Iceland screening has greatly affected the rate of all stages of squamous cell carcinoma, but not the rate of adeno- and adenosquamous carcinomas. In fact the rate of adenocarcinoma has been increasing. CONCLUSIONS: Organized screening is more effective than spontaneous screening in reducing the risk of cervical cancer. Although differences in environmental, biological and ethnic factors may call for different screening strategies, screening should preferably start soon after age 20 with a screening interval of 2-3 years.

Trends in cervical intra-epithelial neoplasia in Iceland through 1995: evaluation of targeted age groups and screening intervals.

BACKGROUND: Targeted age groups and screening intervals are dependent on the age-specific prevalence of the preclinical disease and the length of the detectable pre-clinical phase. This study evaluates the UICC and EC recommendations regarding targeted age group and screening intervals based on the Icelandic screening experience. METHODS: The trends for cytologic preinvasive lesions were analyzed at first visit during the period 1966-1995, at second and later visits after a normal test(s) taken after 1985, and finally at any visit in 1991-1995. The frequency of histologic lesions was calculated for the birth cohort from the age of 60 and among women referred for colposcopic examination in 1994. RESULTS: The prevalence of preinvasive disease has increased significantly since 1980, and the rate of moderate to high-grade cytologic and histologic lesions begins increasing as early as 20 years of age. The rate of these lesions starts to accumulate at 24 to 36 months after a normal smear, but the rate decreases with the number of negative smears taken. Among correctly screened women the rate of histologically verified moderate to high-grade lesions and invasive disease is practically non-existent after the age of 60, while among the younger women cases with microinvasive disease start to appear within 2 to 3 years after a normal smear. CONCLUSION: Screening should start soon after age 20 with a screening interval of 2 to 3 years. The screening interval can probably be extended to 4 years at the age of 50 and stop at the age of 60 to 64 among regularly screened women.

[The Nordic cervical screening programmes through 1995. Evaluation of incidence and mortality rates, targeted age groups and screening intervals.].

BACKGROUND: The objective of cervical cancer screening is to lower the incidence and mortality rates of the disease. The Nordic countries have a long experience in cervical cancer screening. Based on this experience this study evaluates the UICC and EC guidelines regarding the effectiveness of organized vs. spontaneous screening, targeted age groups and screening intervals. MATERIAL AND METHODS: The study analyses the features of the Icelandic and the Nordic screening programmes and the observed trends in the incidence and mortality rates in these countries through 1995. It also analyses the trends for cytologic preinvasive lesions at first visit and at second and later visits after a normal test(s). The frequency of histologic lesions was calculated for the birth cohort 1920-1926 from the age of 60 and among women referred for colposcopic examination in 1994. RESULTS: Organized screening started in all the Nordic countries except Norway soon after 1960. Up to 1985 the target age group and screening interval were most intensive in Iceland. All countries intensified the screening intervals after 1985. The reduction in both the mortality and the incidence rates was greatest in Iceland and Finland, intermediate in Sweden and Denmark, and lowest in Norway but in that country organized screening started in 1994. The age-specific incidence in the 20-29 age group has been increasing since 1971 in all the Nordic countries, except in Finland. In Iceland screening has greatly affected the rate of all stages of squamous cell carcinoma, but not the rate of adeno-and adenosquamous carcinomas. The prevalence of preinvasive disease has increased significantly since 1980. The rate of moderate to high-grade cytologic changes begins increasing as early as at 20 years of age and moderate to high-grade histologic lesions start to accumulate at 24 to 36 months after a normal smear. The rate of these lesions decreases with the number of negative smears taken. Moderate to high-grade histologic lesions and invasive cancer are practically non-existent after the age of 60 among correctly screened women. A strong correlation is found between increased attendance rates and the proportion of cases diagnosed with a Pap smear at stages IA and IB occult. The latter cases mainly occur among women under the Hge of 45 and start to appear less than two years after a normal smear. CONCLUSIONS: Well-organised screening is more effective than spontaneous screening in reducing the risk of cervical cancer. Screening should preferably start soon after age 20 with a screening interval of two to three years. The screening interval can pro-bably be extended to four years at the age of 50 and screening could stop at the age of 60 to 64 among regularly screened women.

Survival and prognostic factors of endometrial cancer patients in Iceland 1964-1985: can attendance at population-based Pap-smear screening affect survival?

After histological review of all cases registered during the period 1964-1985 at the Cancer Registry, 260 cases with endometrial carcinoma were eligible for analyses of survival rates and prognostic factors, as well as the association of Pap-smear screening attendance with these factors and survival. The total age-adjusted 5- and 10-year relative survival rates were 76% and 75%, respectively. The prognostic factors were tested by univariate analysis and simultaneously by a multivariate analysis using the Cox proportional hazards model. Factors that independently gave a less favorable prognosis were non-attendance at screening, older age at diagnosis, deep myometrial invasion, advancing stages and tumor grading, radiotherapy only, extra-genital symptoms and histology types of serous, clear cell and undifferentiated tumors (histologic type 3). Tested simultaneously with the Cox proportional hazards model, parameters that maintained a less favourable prognosis were grade 3, stage III-IV, deep myometrial invasion, older age, radiotherapy only and extra-genital symptoms. In addition, screening attendance showed significant interaction with age. In stages III and IV only grade 3 maintained a significantly less favorable prognosis. We conclude that our results indicate that attendance at Pap-smear screening (taking Pap smears and screening for genital symptoms) has a favorable prognostic value, especially among women under the age of 62.

Human papillomavirus (HPV) in an Icelandic population: the role of HPV DNA testing based on hybrid capture and PCR assays among women with screen-detected abnormal Pap smears.

This study was based on 358 cases with abnormal smears referred for colposcopy and HPV DNA testing. We analysed: 1) the frequency of different grades of cyto- and histopathologic findings; 2) the frequency and relative amount of HPV DNA with the hybrid capture assay (HCA) in swabs, and the frequency of HPV with PCR in swabs (-S) and biopsies (-B); and 3) the frequency of HPV types according to the grade of the cyto- and histopathologic findings. Of all cases, 95% were positive with all HPV tests combined. The HCA (HPV: 16, 18, 31, 33, 35, 45, 51, 52 and 56) and the PCR-S and PCR-B (HPV: 16, 18, 31, 33 and 35) tests for high-risk HPV exhibited sensitivities of 57%, 56% and 48%, respectively. The high-grade smears and the high-risk PCR-S HPV had about 80% sensitivity for histologic high-grade lesions compared with around 70% for HCA and the PCR-B. Combining the high-grade smears and the high-risk HPV increased the sensitivity to 93-96%. Among the cervical intraepithelial neoplasia I (CIN I) and the atypical squamous cells of undetermined significance (ASCUS) smears the sensitivity of high-risk HPV for high-grade histologic lesions was 63% for the HCA and 79% for the PCR-S. No correlation was found between the relative amount of HPV DNA detected by HCA and the grade of cyto- and histological lesions. We conclude that the results strongly indicate that HCA is less sensitive than PCR in the diagnosis of high-risk HPV, that swabs are more sensitive than biopsies as a sampling method, that high-risk HPV and high-grade smears are complementary for the diagnosis of high-grade histologic lesions and that the present role of HPV testing in screening could be limited to identifying women with low-grade smears and koilocytotic or low-grade colposcopic biopsies that are at risk of concealing or developing high-grade histologic lesions.

What attendance rate can be achieved for Pap smear screening? A case-control study of the characteristics of non-attenders and results of reminder efforts.

OBJECTIVE: To understand participation failures in a national Pap smear screening programme by studying characteristics of non-attenders and results of further reminder efforts. DESIGN: A case-control and an intervention study. SETTING: The community health centre in the town of Hafnarfjördur, Iceland. SUBJECTS: The target population comprised 2510 women aged 35-69, who were invited regularly every second year for cervical cancer screening. MAIN RESULTS: 2241 (89.3%) had attended screening during the preceding five years, 102 (4.1%) had never attended, and 167 (6.7%) had attended previously but not during the preceding five years. Women with a mental disorder and those who had never married were more likely not to attend. The most usual explanations given by non-attenders were that they did not like to participate, or they felt they did not need to, some of them because their uterus had been removed. Of the non-attenders 29 (10.8%) came for a Pap smear after repeated reminding efforts. CONCLUSIONS: Total participation rate in cervical cancer screening programmes in Iceland is high. When efforts are taken to lower the non-attendance rate it has to be kept in mind that many women are unwilling or unable to participate in such preventive measures.

Quality assurance in cervical cancer screening: the Icelandic experience 1964-1993.

Monitoring of the effectiveness of a screening programme is vital to ensure optimal use of public resources. This report correlates the results of the Icelandic cervical cancer screening programme with the results of monitoring the programme since 1964. Screening has significantly decreased both the incidence and mortality rates and greatly affected the stage distribution of squamous cell carcinomas, but not of adeno- and adenosquamous carcinomas. In the 25-64 years age group, 84% were screened, 80% of whom were in the organised screening. Smears taken outside the guidelines amounted to 10%. Sensitivity at 1 year was 93% for all smears. At 3 years it was 81% for squamous cell carcinomas, and 42% for adeno-and adenosquamous carcinomas. The rate of unsatisfactory smears was 1.3% for all smears, and 4.5% of the women had abnormal smears (7.7% in the 20-24 years age group). The specificity of the smears test was 98%. It is concluded that monitoring is vital for optimal screening results and although screening is effective in the targeted age group of 25-64 years it should preferably start sooner after age 20 years with a screening interval of 2-3 years.

Effect of organized screening on the risk of cervical cancer. Evaluation of screening activity in Iceland, 1964-1991.

The Icelandic Cancer Society launched a screening programme for cervical cancer in June 1964. The aim was to lower the incidence and mortality rates by screening the age group 25-69 at 2- to 3-year intervals. This report analyses the trends in invasive and pre-invasive disease and the distribution of stage and histology, and also evaluates the attendance, the target age group and the screening interval. Before screening, the incidence and mortality rates were on the increase but both have since fallen significantly. Screening greatly affected the rate of microinvasive and stage > or = II squamous-cell carcinomas but not the rate of adeno- and adenosquamous carcinoma. The mean age at detection of invasive cancer has decreased significantly and cancer has become practically non-existent among correctly screened subjects over the age of 69. Among younger women the rates of moderate and severe pre-invasive lesions at first visit increased significantly after 1980. The rate of these lesions was fairly consistently high only 1 year after the first normal visit. It is concluded that organized screening, co-ordinated with spontaneous activity, is effective in reducing the risk of cervical cancer. Regular high attendance and strict follow-up of abnormal cases is a prerequisite for good screening results. Screening should preferably start at about the age of 20 and extend to 60-69 years of age, depending on the number of negative smears by that age. Screening can safely start with a screening interval of 2 to 3 years, but this interval can probably be extended to 4 or 5 years at older ages.

Screening as a prognostic factor in cervical cancer: analysis of survival and prognostic factors based on Icelandic population data, 1964-1988.

Survival rates were computed for 376 women diagnosed with carcinoma of the uterine cervix between 1964 and 1988. The 5-year survival rate for the entire group was 63%. The effect of age at diagnosis, clinical stage, histopathology, year of diagnosis, and screening program attendance was studied by univariate analysis and simultaneously with a multivariate analysis, the Cox proportional hazards model. All these parameters had a significant effect on survival, with clinical stage as the strongest parameter followed by histology, year of diagnosis, age at diagnosis, and attendance at screening. Women who had attended the cervical screening program fared significantly better than those who had never attended. Patients treated in the late years of the study period had a significantly better survival rate, possibly indicating improved treatment. Young women had a significantly better prognosis than older women. Women with adenocarcinoma and anaplastic tumors had a significantly worse prognosis than women with squamous and adenosquamous carcinoma. The prognostic effect of screening was mainly attributed to the more favorable distribution of early stages and younger age at diagnosis among the screened women. After all the analyzed parameters had been adjusted for the nonattenders still had poorer prognosis.

Trends in cervical and breast cancer in Iceland. A statistical evaluation of trends in incidence and mortality for the period 1955-1989, their relation to screening and prediction to the year 2000.

The time trends in incidence and mortality from cervical cancer and breast cancer in Iceland, from 1955 to 1989, were analyzed by fitting curvilinear regressions to the age-standardized rates. The effect of the screening was evaluated by comparing the curvature of the fitted regression lines and changes in screening activity. The incidence and mortality rates for both cancer types were predicted up to the year 2000. At the commencement of cervical cancer screening in 1964, both the incidence and mortality rates were on the increase. After 1970, both rates decreased significantly. Assuming that regular attendance at screening will be 85%, it is predicted that the incidence and mortality rates will level out at about 7.5 and 2 cases per 100,000 women per year, respectively, by the year 1995 and remain at that level. The incidence of breast cancer has increased steadily since 1955. A sharp rise has been observed since 1987, due to screening with mammography. The mortality rate has shown small but significant fluctuations with time. The incidence rate is predicted to increase at the same rate as before 1987 (i.e. at 1.1 cases per 100,000 women per year), but at a slightly higher level and is predicted to reach 84 cases per 100,000 women per year by the year 2000. Breast cancer mortality is predicted to decrease to about 17 cases per 100,000 women per year by 1995 and to remain at that level.

Catecholamine measurements in urine by high-performance liquid chromatography with amperometric detection--comparison with an autoanalyser fluorescence method.

In order to validate different methods of measuring urinary catecholamines (norepinephrine, epinephrine and dopamine) in humans, methods based on separation of catecholamines using reversed-phase or cation-exchange high-performance liquid chromatography with electrochemical detection were compared with an autoanalyser-based fluorescence method. Different methods for pre-chromatography sample purification were also studied. For measurements of urinary catecholamines, the reversed-phase-based chromatographic techniques studied were found to give less reliable results than cation-exchange chromatography, even if one of them (Clin Rep Urine Catecholamine Kit) gave almost as precise estimates. The autoanalyser technique yielded good results. It is concluded that cation-exchange chromatography with an appropriate sample work-up procedure (a combination of organic solvent extraction and alumina adsorption) is a reliable and accurate method for analyses of urinary catecholamines.

Radiation dose and breast cancer risk in patients treated for cancer of the cervix.

The relationship between breast cancer and radiation treatment for cervical cancer was evaluated in an international study of 953 women who subsequently developed breast cancer and 1,806 matched controls. Radiation doses to the breast (average 0.31 Gy) and ovaries (average 32 Gy) were reconstructed for exposed subjects on the basis of their original radiotherapy records. Overall, 88% of the breast cancer cases and 89% of the controls received radiation treatment [relative risk (RR) = 0.88; 95% confidence interval (CI) = 0.7-1.2]. Among women with intact ovaries (561 cases, 1,037 controls), radiotherapy was linked to a significant 35% reduction in breast cancer risk, attributable in all likelihood to the cessation of ovarian function. Ovarian doses of 6 Gy were sufficient to reduce breast cancer risk but larger doses did not reduce risk further. This saturation-type response is probably due to the killing of a critical number of ovarian cells. Cervical cancer patients without ovaries (145 cases, 284 controls) were analyzed separately because such women are at especially low natural risk for breast cancer development. In theory, any effect of low-dose breast exposure, received incidentally during treatment for cervical cancer, should be more readily detectable. Among women without ovaries, there was a slight increase in breast cancer risk (RR = 1.07; 95% CI = 0.6-2.0), and a suggestion of a dose response with the RR being 1.0, 0.7, 1.5 and 3.1 for breast doses of 0, 0.01-0.24, 0.25-0.49 and 0.50+ Gy, respectively. However, this trend of increasing RR was not statistically significant. If low-dose radiation increases the risk of breast cancer among women over age 40 years, it appears that the risk is much lower than would be predicted from studies of younger women exposed to higher doses.

The value of screening as an approach to cervical cancer control in Iceland, 1964-1986.

The effect of screening for cervical cancer on time trends in incidence and mortality from that disease, and the occurrence of pre-invasive cervical lesions during the period 1964-1986, were analyzed. After commencement of screening in 1964 all the above parameters increased for a short initial period but then fell markedly. From 1980, coinciding with a sharp rise in regular attendance rate, there was an increase in incidence up to 1984, followed by a decrease. The rate of pre-invasive stages also increased from 1980, but appears to be levelling off. The cervical cancer mortality rate decreased significantly during the study period. In more recent years, a shift in the occurrence of cervical cancer and pre-invasive lesions from older to younger women has been observed. Screening still appears to be effective in the control of squamous-cell carcinomas of stages I B and over, but not of adeno- and adenosquamous carcinomas.

Radiation dose and second cancer risk in patients treated for cancer of the cervix.

The risk of cancer associated with a broad range of organ doses was estimated in an international study of women with cervical cancer. Among 150,000 patients reported to one of 19 population-based cancer registries or treated in any of 20 oncology clinics, 4188 women with second cancers and 6880 matched controls were selected for detailed study. Radiation doses for selected organs were reconstructed for each patient on the basis of her original radiotherapy records. Very high doses, on the order of several hundred gray, were found to increase the risk of cancers of the bladder [relative risk (RR) = 4.0], rectum (RR = 1.8), vagina (RR = 2.7), and possibly bone (RR = 1.3), uterine corpus (RR = 1.3), cecum (RR = 1.5), and non-Hodgkin's lymphoma (RR = 2.5). For all female genital cancers taken together, a sharp dose-response gradient was observed, reaching fivefold for doses more than 150 Gy. Several gray increased the risk of stomach cancer (RR = 2.1) and leukemia (RR = 2.0). Although cancer of the pancreas was elevated, there was no evidence of a dose-dependent risk. Cancer of the kidney was significantly increased among 15-year survivors. A nonsignificant twofold risk of radiogenic thyroid cancer was observed following an average dose of only 0.11 Gy. Breast cancer was not increased overall, despite an average dose of 0.31 Gy and 953 cases available for evaluation (RR = 0.9); there was, however, a weak suggestion of a dose response among women whose ovaries had been surgically removed. Doses greater than 6 Gy to the ovaries reduced breast cancer risk by 44%. A significant deficit of ovarian cancer was observed within 5 years of radiotherapy; in contrast, a dose response was suggested among 10-year survivors. Radiation was not found to increase the overall risk of cancers of the small intestine, colon, ovary, vulva, connective tissue, breast, Hodgkin's disease, multiple myeloma, or chronic lymphocytic leukemia. For most cancers associated with radiation, risks were highest among long-term survivors and appeared concentrated among women irradiated at relatively younger ages.