Japanese Journal of Clinical Oncology 30:322-331 (2000)
© 2000 Foundation for Promotion of Cancer Research
Report of the Thirteenth International Symposium of the Foundation for Promotion of Cancer Research: Cancer Screening—Past, Present and Future
1National Cancer Center, Tokyo, Japan, 2University of Tampere, Tampere, Finland and 3Tohoku University School of Medicine, Sendai, Japan
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION OPENINGSESSION 1SESSION 2: ORGAN-SPECIFIC...SESSION 3: PERSPECTIVES OF...SUMMARY OF SYMPOSIUMREFERENCES |
|---|
The Thirteenth International Symposium of the Foundation for the Promotion of Cancer Research was entitled ‘Cancer Screening—Past, Present and Future’ and was held in Tokyo on April 26–28, 2000. The symposium was organized by Drs Tadao Kakizoe, Matti Hakama, Shigeru Hisamichi and Takeshi Sano with Dr Takashi Sugimura as advisor.
|
OPENING |
|---|
|
TOPINTRODUCTIONOPENINGSESSION 1SESSION 2: ORGAN-SPECIFIC...SESSION 3: PERSPECTIVES OF...SUMMARY OF SYMPOSIUMREFERENCES |
|---|
Chairperson: Dr Shigeru Hisamichi
Dr Sugimura (National Cancer Center, Tokyo) opened the symposium with a welcome address and a review of the previous symposia, which have looked at fundamental and clinical research in organ-specific cancer and his personal view of cancer screening (1,2). Since 1981, cancer has been the leading cause of death in Japan and there are many questions relating to early detection and treatment. The role of multistage carcinogenesis has been raised, leading to a progression from normal tissue to precancerous lesions and early cancer with its relationship to the timing of detection (3).
Dr Kakizoe (NCC, Tokyo), in the opening address, raised the issues of the mechanisms of cancer detection. These included presentation with symptoms or following self-check examination, both with uncertainty as to the stage of detection and presentation in the asymptomatic stage. It is the latter which was the brief of this symposium, to promote early detection and hence early and most effective treatment. Also included was the future prediction of cancer development, with detection through genetic studies such as the BRCA 1 and 2 genes in breast cancer, which would allow careful follow-up, chemoprevention and prophylactic treatment.1
|
|
SESSION 1 |
|---|
|
TOPINTRODUCTIONOPENINGSESSION 1SESSION 2: ORGAN-SPECIFIC...SESSION 3: PERSPECTIVES OF...SUMMARY OF SYMPOSIUMREFERENCES |
|---|
1. General View
Chairperson: Dr Shigeru Hisamichi
Dr Matti Hakama (Tampere School of Public Health, Tampere, Finland) in his lecture entitled ‘Cancer screening past, present and future’ gave an overall review of cancer screening with emphasis on the evaluation of the effectiveness of screening programmes. The purpose of screening for cancer is to reduce mortality from the disease by detecting disease at an early stage. Although screening programmes, such as for cervical cancer, have been set up for almost 50 years in some instances and have been shown to confer an effective advantage in terms of risk of cancer, many have not been evaluated according to the stringent rules of evidence-based medicine such as randomized controlled trials. More recent randomized trials have looked at the effectiveness of screening for breast and colorectal cancer but trials for other particular sites lack rigorous design and the effects have probably never been proven or have led to uncertainty. Two main avenues were indicated for future research. These were efforts to improve screening technology and to evaluate screening programmes not only regarding the effect on mortality but also on quality of life and economic burden. Details were presented of the effectiveness of screening in Nordic countries for cervical, breast and colorectal cancer. It has been calculated that this has prevented 4000 deaths in a total 50 000 per year. There would be a total of 25 000 life years gained as a result of the screening programmes but questions are raised as to the cost benefits.
Dr Suketami Tominaga (Aichi Cancer Center Research Institute, Nagoya) in his lecture entitled ‘An overview of cancer prevention in Japan’ outlined the role of screening from the Japanese perspective. Cancer prevention consists of primary cancer prevention (prevention of the occurrence of disease and precursor forms) and secondary cancer prevention (prevention of cancer death by early detection or screening). Trends in mortality rates have shown that since 1981, cancer is the leading cause of death in Japan. This is due to the prevention, early detection and treatment of TB and cerebrovascular disease which results from mass screening programmes set up in the past. Now efforts are being directed towards the screening of cancer. This started with screening for stomach and cervical cancer (1966–67) beginning in the 1960s introduced by the Ministry of Health and Welfare. The mortality rates for these sites has decreased as a result and led to the introduction of breast and lung screening (1987) and colorectal screening (1992). In 1995 these diseases were responsible for 53% of all cancer deaths. Figures were presented which demonstrated the detection rates of the screened diseases and the trend in cancer deaths between 1965 and 1995. It was pointed out that the effectiveness of the screening was analysed in Japan by case-control studies as opposed to randomized controlled trials. More recently there have been efforts to focus on the primary prevention of cancer and improvements in state of the art treatments. These have included improvements in daily habits such as smoking, drinking and diet, elimination of carcinogens from the environment such as pollution and food additives, prevention of tumour-associated vectors such as HBV, HCV and H. pylori and chemoprevention. Chemoprevention of cancer is a promising avenue for the future.
2. Evaluation of Screening Effectiveness
2.1. Methodology to evaluate cancer screening
Chairperson: Dr Gilbert Friedell
Dr Noel S. Weiss (University of Washington, Seattle, USA) in a lecture entitled ‘Study design considerations in seeking to evaluate the efficacy of screening for cutaneous melanoma’ outlined a melanoma screening programme as background for a case-control study used to determine the effectiveness of screening. Cutaneous melanoma varies in incidence geographically, but is prevalent in Australia and New Zealand with 40–50 cases per 10 000 per year. It is particularly amenable to screening since it is visible to the naked eye and many lesions have a lengthy pre-invasive phase. There are two components to evaluating a screening process, these being, the ability to detect lesions early and the effectiveness to treat lesions early. It is difficult to evaluate these in isolation. A case-control study in Australia and New Zealand was presented which was initiated to assess the efficacy of self- and physician screening of the skin. The difficulty of obtaining accurate information on screening history was discussed as well as the design features of the study. New incident cases were interviewed, but analysis was based upon those who go on to die of the disease. Follow-up of incident cases was long enough to identify the majority of melanoma deaths and screening histories were restricted to a fixed date prior to diagnosis. Difficulties with design were introduced as a minimal level of non-response among incident cases and too short a follow-up duration, which may affect validity. Also discussed was the nature of the screening examination in terms of the nature of the activity and the person conducting the test. These various factors were all discussed with reference to the above case-control study with the implications to obtaining a valid estimate of the efficacy of screening for melanoma.
Dr Hiroyuki Shimizu (Gifu University, Gifu) in his lecture entitled ‘Psychological aspects of mass screening for cancer’ reviewed studies which addressed the issues of psychological impact. It was suggested that the psychological effects of a screening programme should also be considered in addition to the obvious benefits of the programme in decreasing cancer deaths. A study reviewing this was carried out among patients in a breast screening programme at Gifu in Japan. Study subjects who included 379 screened patients and 51 examined patients were asked to bid for the value (in yen) attached to a willingness to pay (WTP) to undergo breast screening and have a negative result and a willingness to accept anxiety (WTA) while awaiting a physician’s diagnosis. In all cases the values of WTP exceeded those of WTA. In addition, a survey looked at the maximum permissible limit for a false-negative detection rate and a false-positive detection rate for breast cancer screening among participants and physicians. Data showed that the participants more than physicians expected a high diagnostic sensitivity even for screening tests. It was suggested that a more practical evaluation for cancer screening should be designed to incorporate psychological factors.
2.2 Cost effectiveness of cancer screening
Chairperson: Dr Noel Weiss
Dr Judith L. Wagner (Congressional Budget Office, Washington, USA) in a lecture entitled ‘Cost effectiveness of cancer screening’ discussed the economic factors in investment in screening programmes. In healthcare in many countries, considerations of costs have guided decisions, particularly in prevention. In addition to providing data on the cost of a screening programme to society, the analysis of cost effectiveness can also allow the evaluation of the efficiency of the screening programme and raise key questions. These include (1) who should be screened, (2) how much and what kinds of efforts should be made to encourage individuals to be screened, (3) at what age or ages screening should occur, (4) what screening technologies should be used, (5) what settings of care should screening take place and (6) what follow-up protocols should be used for positive or suspicious results. Cost-effective analysis is used to measure the return on the investment, in terms of improvement in health status, and is used in government before the instigation of programmes in order to make choices. Two examples were presented of cervical and colorectal cancer to illustrate the different methods and the effect of these on cost–benefit analysis. One showed a substantial improvement in effect at a modest increase in cost and the other a small improvement in effect for a major increase in cost. Analyses have shown benefits between either screening or no screening. The analysis between screening models and technologies is less certain and the constant development of new technology makes it difficult to identify the best screening method for a particular disease.
Dr Ichiro Tsiji (Tohoku University, Sendai) in his lecture entitled ‘Cost effectiveness of cancer screening in Japan’ reviewed the economics of cancer screening in Japan. At present, 5% of the gross national product of Japan is spent on national medical expenditure, as opposed to 10% in USA. The proportion spent on medical expenditure is increasing as against the average national earnings. In 1997, there were 59 million screening procedures in the five national screening programmes (stomach, uterus, breast, lung and colorectal), costing a total of US$1.93 billion, equivalent to 0.7% of the total medical budget in Japan. The cost effectiveness of the five screening programmes was analysed, comparing one-time screening with no screening, with effectiveness defined as number of life years saved. Costs included those for screening and diagnosis, cancer treatment and terminal care. The cost-effective ratio was calculated (additional cost of screening/additional number of life years) for stomach cancer (US$1470 in men and US$7070 in women), endometrial cancer (US$10 470), breast cancer (US$24 600), colorectal cancer (US$60 in men and US$1330 in women) and lung cancer (US$6230 in men and US$21 100 in women). The results presented show that colon cancer screening is the most effective with a cost saving in screening documented, and lung cancer screening the least effective. Also, analysis of breast cancer screening has shown a greater cost effectiveness with screening by biennial mammography as opposed to clinical examination only. This point was debated, particularly in view of recently published analyses of randomized trials of breast screening.
|
SESSION 2: ORGAN-SPECIFIC ASPECTS |
|---|
|
TOPINTRODUCTIONOPENINGSESSION 1SESSION 2: ORGAN-SPECIFIC...SESSION 3: PERSPECTIVES OF...SUMMARY OF SYMPOSIUMREFERENCES |
|---|
1. Colorectal Cancer Screening
Chairperson: Dr Harald zur Hausen
Dr Joe V. Selby (Kaiser Permanente, Oakland, USA) in his lecture entitled ‘Colorectal cancer screening 2000—so many choices’ outlined the evidence for the effectiveness of screening for colorectal cancer, which shows it to be an ideal target. Screening both shifts the stage distribution of disease towards earlier lesions and allows effective treatment at a pre-malignant stage, so reducing the incidence in the screened population. Evidence was presented for effective screening in the form of three case-control studies involving the use of faecal occult blood testing, three case-control studies involving the use of rigid and flexible sigmoidoscopy, two cohort studies which demonstrated the benefit of endoscopic polypectomy of pre-malignant lesions and three randomized trials showing the benefit of screening with FOB tests in reduction of mortality from colorectal cancer. A screening programme with sigmoidoscopic examination based in Northern California at the Kaiser Permanente Institute was described which included 118 000 examinations (65% of the population). There were in this series six hospitalized complications, which included perforation, bleeding and diverticulitis following examination, but the data indicated the patients’ willingness to undergo screening examination and demonstrated a decline in mortality in colorectal cancer as a result of the programme. Patient reluctance to undergo colorectal cancer screening does not explain the low rates of screening reported elsewhere and the fact that screening for colorectal cancer is not widely adopted. The most challenging issues relate to the optimum screening regimen, with a wide choice of alternative methods. In making these choices, cost effectiveness and compliance must be taken into consideration. Colonoscopy is the most sensitive test for colorectal screening available, but the optimum strategy must be worked out in terms of the financial constraints of setting up the programme and training personnel. Sigmoidoscopy is highly effective in screening for distal tumours but leaves the proximal colon unprotected. The addition of FOB may be a reasonable approach, particularly as the screening test may be carried out by nurses, thus freeing up physicians. The cost benefits are yet to be calculated.
Dr Hiroshi Saito (Hirosaki University, Hirosaki) in his lecture entitled ‘Screening for colorectal cancer by immunological faecal occult blood testing’ described a more specific test to replace FOB tests. Randomized controlled trials have shown the benefit in terms of a reduction in mortality from colorectal cancer in patients screened using a guaiac-based FOB test. This test, however, is relatively poor in terms of sensitivity and specificity. An immunological FOB test using a reversed passive haemagglutination reaction (RPHA) has been shown to have greater validity. The sensitivity for detection of colorectal cancer by this test has been reported to be 67–89% compared with 33–37% for the guaiac test. Similarly, case-control studies have shown that screening reduces the mortality of colorectal cancer by 60% compared with 15–33% with the guaiac test. A screening programme incorporating FOB tests was introduced in Japan in 1992. This is offered to persons over 40 years old as an immunological FOB test with diagnostic investigation with colonoscopy or barium enema for those with positive results. In 1995, 4.3 million people participated in the screening programme with a 7% rate of positive tests, which led to the diagnosis of 4000 cases of invasive cancer. Compliance with diagnostic tests (60%) was noted to be lower than that in other cancer screening programmes. In order to achieve an effective screening programme, a high compliance as well as a sensitive diagnostic test is essential and it was suggested that compliance may be linked to an insufficient knowledge concerning the screening programmes.
2. Prostate Cancer Screening
Chairperson: Dr Tadao Kakizoe
Two papers by Dr George Bartsch (University of Innsbruck, Innsbruck, Austria) were discussed in absentia. The first, entitled ‘Decrease in prostate cancer mortality following introduction of prostate specific antigen (PSA) screening in the federal state of Tyrol, Austria’ supported the use of screening tests for the early detection of prostate cancer. The study, which began in 1993, incorporated a freely available PSA test to the population of males aged 45–75 years. This showed an increase in the number of potentially curable cancers with significant stage migration to early lesions and a reduction in the mortality due to prostate cancer. The second paper, entitled ‘Prostate cancer screening in Tyrol, Austria: experience and results’, suggested that PSA-based screening with low PSA cut-off values (2.5 ng/ml in men aged 45–49 years and 3.5 ng/ml in men aged 50–59 years) increased the detection rate of clinically significant, organ-confined and potentially curable prostate cancer. In this study, all patients with PSA levels <4 ng/ml underwent curable radical prostatectomy. The percentage free PSA and PSA transition zone density provide an additional benefit over total PSA. Discussion of these papers, which showed the decrease in mortality in prostate cancer in this region of Austria, centred around the possibility of local improvements in the treatment in this region over the time period; however, owing to the slow progression of the disease, the full impact of the screening programme may not be evident for 10–15 years.
Dr Hiroki Watanabe (Meiji University of Oriental Medicine, Kyoto) in his lecture entitled ‘Mass screening of prostate cancer in Japan’ outlined the screening for this disease in Japan with reference to its lower prevalence in Asia compared with the West. The first trial for screening of prostate cancer in Japan involved the use of transrectal ultrasound; however, with the discovery of PSA in 1979, PSA screening has been introduced with an improved detection rate. In 1994 proposals were made as to the optimum screening programme in Japan which included PSA screening in males over the age of 55 years. The ranges of PSA cut-offs were as follows: PSA <4 ng/ml, negative (93% population; however, there is a 0.2% false negative), PSA 4–10 ng/ml, grey zone (5% population) and PSA >10 ng/ml (2% population, 40% of cases have prostate cancer). Patients in the last group were directly referred to urologists for biopsy; however, those in the ‘grey zone’ are selected for secondary survey with either digital rectal examination, PSA density and F/T ratio or prostatic six-sector biopsy under ultrasound guidance. The total detection rate for prostatic cancer by this system is 1.3%, which reflects the lower prevalence of the disease compared with the west. Data were presented which indicated a reduction in the mortality in prostatic cancer in those regions which are screened compared with those which are not. Also discussed was new technology used in the decision to biopsy those patient belonging to the ‘grey zone’. Although a six-segment biopsy under ultrasound guidance is the global standard, a method for increasing the sensitivity as well as reducing the number of biopsies taken was shown using power Doppler imaging. This used blood flow to direct the biopsy needle into hypervascular lesions with an increased sensitivity of 80% over conventional ultrasound.
3. Breast Cancer Screening
Chairperson: Dr Won Chul Lee
Dr Edward Hendrick (Lynn Sage Comprehensive Breast Center, Chicago, USA) in his lecture entitled ‘Quality assurance in breast cancer screening’ described systems in place to ensure quality assurance in mammographic breast cancer screening. Screening for breast cancer has been indicated as having benefits in terms of decreasing mortality. In order to realize the full benefits of the screening programme, with consistently high standards, it is practice to consider quality control of the technical aspects of the screening mechanism and equipment and quality assurance, including checks of film interpretation and the quality of follow-up. Certain goals of medical audit are established with values set for parameters of recall rate, biopsy rate and positive biopsy rate, proportion of cancers which are screen detected as a fraction of the total and proportion of early lesions detected. The American College of Radiology has set up a system to ensure the uniformity of reporting of mammograms with the results correlating to the probability of malignancy. The international breast cancer screening network set up a study of quality assurance mechanisms in national breast cancer screening programmes. A comprehensive questionnaire was sent to 23 countries including questions on quality assurance activities, requirements, organization and outcomes. A response was obtained in 22 cases with all but one requiring quality control mechanisms with variation of the personnel responsible. The conclusions were that practices ensuring quality assurance were well developed in most instances with the majority following defined protocols with continual performance monitored with the audit requirements outlined above.
Dr Noriaki Ohuchi (Tohoku University, Sendai) in his lecture entitled ‘Quality control of mammographic screening for breast cancer in Japan’ described the recent changes proposed for the Japanese screening programme. In Japan breast cancer has superseded stomach cancer as the leading incident cancer and the highest mortality in women. The mortality is increasing with rates in 1998 of 13.4 per 100 000, nearly four times that reported in 1955. In 1994 clinical breast examination was introduced as a screening modality. A case-control study has recently shown that this alone lacks effectiveness in asymptomatic patients, although it is effective in those who are symptomatic. There are plans to introduce mammography in breast screening programmes this year. The programme has been designed to incorporate sensitivity and cost-effectiveness data from the experience gathered in the trial in Miyagi prefecture. This pilot study has been running since 1989 with screening of 50–69-year-old women (and recently 40–69-year-old women) by biennial mammography and clinical breast examination. Initial studies have shown a detection rate of 3–4 times that for clinical breast examination alone and the detection of earlier lesions. The recommendations for the national screening programme are for women over 50 years to have biennial mammography and clinical breast examination and women 40–49 years old to undergo annual clinical breast examination. An educational programme has also been set up for training personnel and with the production of the Japanese Guidelines for Mammographic Screening. A survey was conducted nationwide to assess the quality control practices at 104 facilities which indicated an improvement in the system over the last decade and uniform high standards in terms of average dose exposure and frequency of phantom images.
4. Lung Cancer Screening
Chairperson: Dr Kunio Doi
Dr Claudia Henschke (Weill Medical College of Cornell University, New York, USA) in her lecture entitled ‘Early lung cancer action project: findings on baseline and annual repeat screening CT’ outlined a project to assess the screening of high-risk patients for lung cancer with low-dose CT scanning. The incidence and mortality rate from lung cancer are high both in the West and in Japan. A randomized controlled trial was described based at the Mayo clinic which used chest X-ray and sputum cytology in a high and a low screened population in order to detect a decrease in mortality in screened individuals from lung cancer. This study showed no benefit in screening for lung cancer by this method and is only one of such trials carried out in the West in the 1970s and 1980s, similarly showing no detectable difference. The detection of lung cancer by chest X-ray required the lesions to be >10 mm (equivalent to 30 doubling times) whereas detection using CT scanning is able to detect lesions up to 2 mm (equivalent to 20 doubling times), leading to much earlier diagnosis. Recently with the introduction of new ultra-fast CT scanners, scan times have decreased to a few seconds with low-dose exposure to radiation. An assessment of the role of CT scanning in screening was carried out on 1000 asymptomatic, medically fit individuals but with a history of smoking. A baseline chest X-ray and CT was carried out on all patients with CT repeated annually. The discovery of non-calcified pulmonary nodules resulted in short-term high-resolution CT with 3D image reconstructions to delineate changes in size. Baseline screening with CT resulted in the diagnosis of lung cancer in 2.7% of cases with 2.3% stage I as opposed to a 0.7% detection rate and 0.4% stage I with chest X-ray alone. Of the detected lesions 96% were resectable and 85% were stage I, with 83% of these lesions not seen on chest X-ray. Annual repeat screening resulted in the diagnosis of a further 36 non-calcified nodules of which seven were malignant on biopsy. It was concluded that annual repeat CT screening for lung cancer was cost effective (US$5000–10 000 per life saved) with the diagnosis of disease at an earlier and more curable stage.
Dr Tomotaka Sobue (NCC, Tokyo) in his lecture entitled ‘Lung cancer screening in Japan’ outlined the present status of lung cancer screening in Japan with reference to studies on the efficacy of chest X-ray and more recently CT. Lung cancer in Japan now ranks as the first leading cause of cancer death for males and the fourth for females. Since 1987, screening by chest X-ray (and sputum cytology for high risk) has been carried out in Japan on all individuals over 40 years old, with 7 million screened annually. Evidence for the efficacy of chest X-ray screening was presented with relatively recent six case-control studies carried out in Japan. All these studies have shown benefits in terms of a reduction in mortality from lung cancer in the screened compared with the unscreened population and based on this evidence the screening programme continues in Japan. Previous randomized studies carried out in the West have failed to show a detectable benefit and the results of a randomized controlled trial in the USA (PLCO trial), which has been ongoing since 1993, are awaited. Although the benefits of screening in the Japanese population are documented by case-control studies, this is offset against the poor survival amongst the screened cancers compared with that of other sites. To this end, other modalities such as CT have been assessed to endeavor to detect lesions at an earlier stage. A report was made of the results of projects which have looked into CT scanning, including the Anti-Lung Cancer Association which carried out screening on 1682 individuals. Amongst these, 36 lung cancer cases were diagnosed, 24 by CT alone, four by chest X-ray and CT, four by sputum cytology alone and four by all three modalities. Those detected by CT were predominantly stage 1A and resectable. The 5-year survival for the whole group was 71% and 87% for those detected by CT alone. This indicated that CT was a valid screening method, but criticisms in the analysis exist with overdiagnosis and lead time bias as to the ultimate effectiveness.
5. Gastric Cancer Screening
Chairperson: Dr Joe V. Selby
Dr Yoshitaka Tsubono (Tohoku University, Sendai) in his lecture entitled ‘Evaluation of screening for gastric cancer with photofluography’ outlined the gastric cancer screening programme in Japan. Mass screening for gastric cancer with barium photofluorography in Japan has been ongoing since 1960. The screening takes place in the community or workplace setting, usually in mobile units. In each screening, seven films are taken by trained radiographic technicians and all films are reported in the cancer centre by two independent physicians. Diagnostic work-up for positive or suspicious tests involves endoscopy, biopsy and surgical referral. On average, 6.4 million individuals are screened annually, with12% undergoing further investigation with an overall detection rate of 0.11%. Based on follow-up studies of the screened population, the sensitivity of barium photofluography is 70–90% and the specificity is 80–90%. The 5-year survival of patients with screen-detected cancer is 15–30% better than that for symptomatic cancer. There have been no randomized controlled trials to assess the effectiveness of screening, but three case-control studies have been consistent in their finding that screening benefits by a decreased risk of mortality from gastric cancer in the screened population. These studies have shown that the risk of mortality decreased by 60% in men and 50% in women. One small cohort study, however, showed no effect. Despite this, screening for gastric cancer by barium photofluography is considered to be cost effective at reducing mortality from this disease.
Won Chul Lee (Catholic University of Korea, Seoul, Korea) in his lecture entitled ‘Current status of stomach cancer screening in Korea’ reported on the setting up recently of a gastric cancer screening programme in Korea. Stomach cancer is the leading incident cancer and the leading cause of cancer death in Korea. Despite this, the screening rate is low and available to a limited number of insured individuals. The participation rate is 1.5% among those eligible and the detection rate is 0.3% for males and 0.1% for females. The principal modality of screening, either barium upper GI series or endoscopy, has not been established but suggestions for screening men over age 40 and women over 45 have been made. Screening for gastric cancer was initiated in 1996 in Korea, along with other programmes of cancer screening. The costs involved in barium upper GI series and endoscopy in Korea are equivocal and some choice is given to those volunteering for screening. The evidence for effectiveness comes from case-control studies, particularly a hospital-based study of 4000 patients which showed a stage shift effect, with an increase in the proportion of stage I lesions diagnosed in the screened population compared with those with symptomatic presentation. There is a need in the near future to provide more data on this population, with studies looking into the sensitivity and specificity of screening tests, cost-effective analysis and measures to increase the compliance of the eligible population
6. Cervical Cancer Screening
Chairperson: Dr Matti Hakama
Dr Harald zur Hausen (Deutsches Krebsforschungszentrum, Heidelberg, Germany) in his lecture entitled ‘The role of HPV detection in screening for cervical cancer’ summarized the theory behind the testing of human papilomavirus (HPV) types in cervical cancer. Approximately 40 different genotypes have been identified so far associated with cervical disease, types 16 and 18 (high risk) associated with CIN and invasive cancer and types 6 and 11 (low risk) associated with condylomata acuminata. In cancer of the cervix, 60–70% of biopsies contain HPV types 16 and 18 and an additional 20% contain types 31, 33, 39, 45, 52, 56, 58 and 66. The virus contains several oncogenes and an explanation was given of the pathogenesis of viral DNA in the modification of host gene expression leading to CIN and invasive cancer. Screening for HPV allows those patients with high-risk infections to undergo regular cytological and colposcopic follow-up. There are a number of test systems available such as serological tests to detect antibodies or the demonstration of viral DNA in lesions. These use expensive and time-consuming techniques and allow only limited testing of a few types of this heterogeneous group. Infection with the high-risk type 16 results in the overexpression of a cellular protein p16 which is involved in the regulation of cell mitosis. The detection of this protein indicates high-risk HPV type 16 infection and serves as a useful marker for CIN and invasive cervical cancer. Demonstration of the use of this marker was given in immunohistochemical stained slides of CIN and cervical cancer showing p16 positivity. In addition, a survey of screening centres for cervical cancer in the USA, UK and Germany showed a discrepancy of only 3% for the detection of the marker p16 as opposed to the detection of cervical dysplasia in standard Papanicolaou smears.
Dr Daisuke Aoki (Keio University, Tokyo) in his lecture entitled ‘Screening programme for cervical cancer in Japan’ outlined the national programme for the screening of both cervical as well as endometrial cancer. The screening for cervical cancer in Japan started in the 1960s and became a national programme in 1983. The mortality rate has decreased from 21.3/100 000 in 1960 to 5.4/100 000 in 1995. Screening is offered to all women over 30 years old and takes the form of a Papanicolaou smear once a year. The screening programme recently incorporated 3.8 million individuals per year, with a detection rate of 0.06%. The screening is carried out by cytopatholgists and technologists authorized by the Japanese Society of Clinical Cytology and reports a false-negative rate of 9.2% and a false-positive rate of 0.55%. The screening of cervical cancer has been shown to be cost effective and to have resulted in a decrease in the incidence and mortality of invasive cervical cancer. The ratio of CIN to invasive cancer is 3:2 in screened cases and 1:4 in symptomatic cases. Debate still continues as to the lower age limit for screening. In Japan, since 1987, screening for endometrial cancer is also offered to those individuals at risk. These criteria are abnormal genital bleeding, age more than 50 years, irregular menstrual bleeding, post-menopausal or nulliparous women. Screening is carried out at the same time as that for cervical cancer by a device to sample the uterine cavity. The incidence of endometrial cancer has been increasing in Japan, but the proportion of early lesions detected through the screening programme is greater with an improved 5-year survival. The effectiveness of the screening programme for endometrial cancer is currently under investigation, with at present no other country screening for this disease.
Dr Alexander Chang (Chinese University of Hong Kong, Hong Kong) in his lecture entitled ‘Overcoming laboratory problems associated with the setting up of a community Pap test clinic: a Hong Kong experience’ gave an account of the automated system involved in Pap screening in Hong Kong. Until recently there was no screening service for cervical cancer in Hong Kong, resulting in it being the fourth ranking malignancy among women, with 450–500 new cases per year. In 1995 a project set up a clinic in Shatin community to provide a free Pap smear to women with no age restriction, information on disease prevention, results by mail and follow-up treatment if necessary. Owing to the labour-intensive screening by cytotechnologists, it was decided to invest in an automative process to speed up laboratory examination. The ‘AutoPap’ screening instrument is used to sort those smears likely to contain abnormal cells. It consists of a high-speed video microscope, digitizer and an algorithm to recognize abnormal cells. Set at a sort rate of 50%, it will divide those slides into those requiring no further review or those requiring detailed examination. In addition, each slide is given a location guided map of 15 fields of view to speed up the process of human screening. A study has confirmed the improvements in accuracy of reporting, as well as solving the problem of shortage of cytotechnicians. The equipment through a pathfinder mapping system was able to check the adequacy of human screening and so maintain quality assurance standards. In the project at Shatin which screened individuals from 18 to 84 years old, at least 100 high-grade lesions have so far been detected.
7. Pancreatic Cancer Screening
Chairperson: Dr Robert A. Smith
Dr Tadao Kakizoe (NCC, Tokyo) in his lecture entitled ‘Screening of pancreas cancer: a challenge’ introduced the possibility of screening for pancreatic cancer in the future. Pancreatic cancer has an aggressive course and poor prognosis, the majority of patients dying of the disease within 1 year of diagnosis. It is increasing in incidence and is now the fifth leading cause of cancer death in Japan with 17 000 deaths in 1997. The only hope of cure remains with surgical resection of early disease, but the majority present with advanced and metastatic condition. In order to improve survival, early detection is a prerequisite. The concept of screening for this disease is challenging and some early initiation of the process was presented. A multi-institutional study was conducted which screened 1608 individuals with a suspicion of underlying disease by ultrasound and CT. Subjects were more than 40 years old and presented to clinics with symptoms such as non-specific abdominal pain, backache, weight loss, icterus and worsening diabetic control or abnormal tests such as amylase, CA19-9 and CEA. From this set, 185 were diagnosed with pancreatic cancer (12%) and of these 15% were stage I and 41% were resectable. The modalities CT and US were complementary in 80% of cases, but 11% were seen only on CT imaging, most of these being resectable lesions in the tail of the pancreas. Imaging of this kind is possible but practical only for a selected group with suspicious criteria for the disease. The search for a suitable marker for screening purposes, present in blood or urine continues, however. Recent studies have detected k-ras mutations in pancreatic juice in those with pancreatic pathology, but again this is not a valid proposition for screening. A selection of serum markers have been studied such as CD44V6, E-selectin, ICAM-1 and ß-hCG, but all of these lack any kind of specificity in pancreatic cancer, being similarly elevated in pancreatitis, pancreatic cyst and a range of other gastrointestinal conditions. They are therefore invalid as markers of pancreatic cancer. The challenge for the future is to be able to screen the population for pancreatic cancer with a new tumour marker to permit its earlier detection.
|
SESSION 3: PERSPECTIVES OF CANCER SCREENING |
|---|
|
TOPINTRODUCTIONOPENINGSESSION 1SESSION 2: ORGAN-SPECIFIC...SESSION 3: PERSPECTIVES OF...SUMMARY OF SYMPOSIUMREFERENCES |
|---|
1. Novel Approach
1.1. Development of screening technology
Chairperson: Dr Claudia Henschke
Dr Kunio Doi (University of Chicago, Chicago, USA) in his lecture entitled ‘Cancers on radiographic images in screening programmes’ introduced the concept of computer-aided diagnosis (CAD). Radiographic images are widely used in screening systems, but accuracy is limited by physicians who may fail to detect lesions in up to 30% of cases. The purpose of CAD is to increase the quality and productivity of reporting by providing a second opinion. The system comprises three major components: image processing for enhancement and extraction of lesions, quantitation of image features and data processing for classifying images into defined patterns. This can help distinguish between normal and abnormal and benign and malignant features. The development of schemes for the detection of masses and microcalcification in mammograms and lung nodules in chest X-rays was described, including a system developed in 1994 in Chicago for mammographic reporting. Observer performance studies have shown that for mammograms, CAD was able to increase the detection accuracy of reporting microcalcification from 80 to 90%. In a study reviewing 12 670 cases, there were 79 breast cancers of which 23 had shown a negative mammogram 1 year before but lesions were visible in retrospect. The CAD detected 12 of these lesions on the previous mammogram. Observer performance studies have shown that for chest X-ray, CAD was able to increase the detection accuracy of reporting lung nodules from 76 to 86%. In a study in Yamaguchi prefecture, there were 39 false-negative cases in a screening programme of which 15 were detected by CAD on the previous films. An alternative method of improving detection rates was described with temporal subtraction imaging to detect changes between two films taken at different times by fitting the radiographs together. This enhances the changes and thus aids the physician in detecting pathological progressive disease. In a study in Iware prefecture, 4780 chest X-rays were reviewed, with previous images to provide a subtraction view. This increased the number of suspicious lesions reported. A view of the role of CAD in the twenty-first century was given showing the development of CAD to allow a comprehensive clinical differential diagnosis and the setting up of mega databases to aid pattern recognition.
Dr Noriyuki Moriyama (NCC, Tokyo) in his lecture entitled ‘Screening of lung cancer by helical CT’ outlined the role of helical CT in the early diagnosis of lung cancer. This procedure takes 15 s with a single breath hold and low-dose radiation equivalent to six plain radiographs. Screening for lung cancer in Japan began in 1975, with chest X-ray and sputum cytology, but the introduction of helical CT into the procedure has increased the detection rate from 0.16 to 0.36%. There is also a stage shift demonstrated in those screened cases by helical CT, with a decrease in the mean size of lesion from 30 to 15 mm, an increase in the proportion of stage I cases from 53 to 83% and an increase in 5-year survival from 38 to 71%. The programme included 10 807 screening examinations, in which 38 cases of lung cancer were diagnosed. Diagnosis was made in eight cases by cytology, in eight cases by chest X-ray and in 34 cases by CT. The importance of sputum cytology was highlighted by the four cases which were diagnosed by this method alone, underlining this essential procedure for squamous carcinoma of the large bronchi which are difficult to visualize by CT. No cases were diagnosed by chest X-ray alone. It was demonstrated that CT is able to diagnose a greater proportion of adenocarcinoma, particularly lesions infiltrating the surface of the alveolae and invisible to plain radiography.
1.2. Tumour markers for screening
Chairperson: Dr Anthony B. Miller
Dr Kazumasa Miki (Toho University, Tokyo) in his lecture entitled ‘Serum pepsinogen as a diagnostic marker of stomach cancer’ outlined the efficacy of serum pepsinogen as a screening test for gastric cancer. Chronic active gastritis is widely accepted as a precancerous condition with a high correlation in Japanese populations between its prevalence and the mortality rates for gastric cancer. Three types of serum pepsinogen types have been identified with PGI reflecting acid secretion by fundic glands. It has been reported that severe chronic active gastritis is associated with a PGI <70 µg/l and a PGI/PGII ratio <3. A study was described in which the serum pepsinogen test was used as a screening modality for gastric cancer in a Tokyo workplace with tests PGI <70 µg/l and a PGI/PGII ratio <3 considered positive. Between 1991 and 1996, 30 899 employees were screened. There were 7708 positive tests (24.9%) and as a result 52 cases of gastric cancer were diagnosed with a detection rate of 0.17%. Of the detected lesions 38 (73%) were early stage and 13 (25%) of these were treated by endoscopic mucosal resection. In the same workplace over the preceding 5 years, screening for gastric cancer had taken place with conventional contrast radiographs. In this period 15 050 employees were screened. There were 2974 positive or suspicious X-rays (19.8%) and as a result 10 cases of gastric cancer were diagnosed with a detection rate of 0.07%. The superiority of the test was demonstrated with a 2.6 times detection rate over conventional screening, a sensitivity of 80%, a specificity of 70% and also a high rate of early lesions detected which were amenable to local endoscopic treatment. The advantages were also cost effective with X-ray detection costing 8.3 million yen per lesion detected and blood tests costing 2.2 million yen. It was proposed that this should in future be adopted for widespead mass screening, but caution was advisable in interpretation in the light of data from studies in Finland on banked serum samples showing a lack of sensitivity and specificity of the tests above. In addition, the results so far displayed may indicate an effect of prevalence round over incidence round screening which would diminish over time.
Dr Ken Yamaguchi (NCC, Tokyo) in his lecture entitled ‘Pro-gastrin-releasing peptide as a specific tumour marker for small cell lung cancer patients’ described the experience with a new tumour marker for a subtype of lung cancer. Small cell lung cancer commonly arises in the lung hilar and hence there is difficulty with diagnosis using conventional imaging. Treatment of this intractable disease is usually with chemoradiation. The tumour is comprised of neuroendocrine cells which secrete a number of peptides, including gastrin-releasing peptide (GRP), and the plasma levels are often elevated in patients with this tumour. The half-life in serum of GRP is only 3 min; however, a cleavage fragment Pro-GRP is more stable and an ELISA assay has been produced to detect this molecule. It was found that Pro-GRP is elevated in 65% of cases of small cell carcinoma of the lung compared with 0.4% of normal subjects, indicating a high specificity, and in addition it is elevated to the same proportion in those with extensive disease (68%) compared wiht those with early disease (57%). Changes in the serum Pro-GRP can be used to follow clinical response and recurrence of disease. It was demonstrated that the high specificity of the test for small cell cancer of the lung produced a 97% certainty of diagnosis in the presence of an abnormal chest X-ray and elevated serum Pro-GRP. The incidence of small cell cancer in male smokers particularly, justifies screening for the disease, but it was pointed out that this was an intractable condition with poor outcome even following treatment.
1.3. Identification of high-risk population by genetic diagnosis
Chairperson: Dr Alexander R. Chang
Dr Robert A. Smith (American Cancer Society, Atlanta, USA) in his lecture entitled ‘Current challenges in the identification and screening of individuals at inherited risk for cancer’ outlined the concept of genetic screening in relation to breast and ovarian disease. The majority of cancers arise due to somatic mutations, but a small proportion are attributable to germline mutations which are passed on as hereditary cancer syndromes. These syndromes are rare in the general population and the features include early age of onset, multiple primary cancers and multifocal or bilateral disease. The disease expression depends on the carrier frequency and the gene penetrance, which is in turn affected by factors such as modifier genes, carcinogens, response to DNA damage and hormonal factors. It has been estimated that between 5 and 10% of cases of cancer are attributable to hereditary cancer syndromes and to date approximately 30 have been identified. The cancer family history, detailing a three-generation pedigree, is the key to effective risk management and counselling. This records information on all individuals relating to age and stage of diagnosis, precursor lesions, outcome and environmental factors, but it is concurred that the history is dynamic and may in some circumstances be unknown or inaccurate. The development of molecular genetics has allowed the emergence of DNA testing of specific germline mutations. This allows the detection of carrier status or the absence of the gene in a specific individual in a high-risk family, which allows the targetting of management issues or support to families. The practicalities of genetic testing were discussed with reference to the BRCA 1 and 2 genes in breast and ovarian cancer. These are tumour suppressor genes on chromosome 17 with high penetrance and result in a lifetime risk of development of 50–85% for breast cancer and 10–20% for ovarian cancer and an increased risk of prostate and laryngeal cancer. The issues of chemoprevention as well as prophylactic surgery were raised in relation to the screening of the BRCA 1 and 2 genes, but there are no follow-up data concerning prophylactic mastectomy or oophorectomy in this condition. The economic, scientific and ethical challenges were discussed with regard to the issue of testing individuals. These included the importance of pre-testing genetic counselling, informed consent, confidentiality, family social issues and decisions regarding surveillance and future management. It was emphasized that genetic susceptibilty was not a general population screening method and consideration of the family pedigree must be made when interpreting the results and estimating the risk for an individual or family.
Dr Shozo Baba (Hamamatsu University, Hamamatsu) in his lecture entitled ‘Identification of high-risk population by genetic diagnosis: insight into colorectal carcinogenesis and its prevention’ outlined the role of genetic testing in familial colorectal cancer syndromes. The development of molecular biology and cancer genetics has allowed the identification of proteins encoded by inherited cancer genes implicated in diverse cellular functions including transmembrane receptors (MET, PTCH, RET), cytoplasmic regulatory or structural proteins (APC, PTEN, NF4), regulators of transcription (p53, Rb1, VHL, WT1), cell cycle factors (CDK4, p16) or DNA damage repair proteins (hMSH2, hMLH1, BRCA1, BRCA2, ATM). The discussion focused on the familial colorectal cancer syndromes familial adenomatous polyposis (FAP) and hereditary non-polyposis colorectal cancer (HNPCC). A registry of 804 such cases in 196 pedigrees has been set up in Japan. In the case of the APC gene activation by ß-catenin-Tcf4 is a critical event in colorectal carcinogenesis leading to adenoma carcinoma progression. This leads to insight into chemoprevention and studies have shown the effect of cyclooxygenase in a mouse model. The effect of Sulindac is in a clinical trial but unlikely to replace colectomy as a preventative measure. In the case of HNPCC, which is autosomal dominant with mismatch DNA repair genes, certain criteria must be met in order to make the diagnosis in the context of Lynch syndromes I and II. A recent study indicates that aspirin may induce apoptosis in such affected cells, but the full role of chemoprevention including surgery in this condition remains to be elucidated.
2. Management of Cancer Screening
Chairperson: Dr Judith L. Wagner
Dr Anthony B. Miller (Deutsches Krebsforschungszentrum, Heidelberg, Germany) in his lecture entitled ‘The US PLCO screening trial’ gave an account of the randomized controlled trial based at the US National Cancer Institute to evaluate the effect on mortality and sensitivity, specificity and positive predictive value of a number of screening tests for cancer of the prostate, lung, colon and ovary. Recruitment commenced in 1994 of eligible individuals, between the ages of 55 and 74 years, without prior history of PLCO cancer, in 10 study centres throughout the USA. It is expected that when recuitment closes in 2010, the enrollment will be 74 000 men and 74 000 women and follow-up will be for 10 years. Randomization follows informed consent and the non-screened population receive the usual care in the community. Screening for prostate cancer will be by annual PSA (PSA >4 µg/ml) and digital rectal examination with a study power of 90% in order to detect a 20% reduction in mortality. Screening for lung cancer will be by annual chest X-ray with a study power of 90% in order to detect a 20% reduction in mortality. Screening for colorectal cancer will be by flexible sigmoidoscopy to 60 cm at 5-year intervals with a study power of 99% in order to detect a 20% reduction in mortality. Screening for ovarian cancer will be by annual trans-vaginal ultrasound and serum CA-125 with a study power of 88% in order to detect a 30% reduction in mortality. A number of ancillary surveys exist in conjunction, in order to provide tissue sample banks and data on epidemiology and aetiology. There was a discussion on the pros and cons of screening programmes in relation to the interpretation of the data from trials, overdiagnosis, the complications of therapy, the prolongation of life and the effectiveness of current treatment.
Dr Gilbert H. Friedell (Markey Cancer Center, Lexington, USA) in his lecture entitled ‘The role of cancer registries in planning and evaluating screening activities’ gave an account of the use of the cancer registry set up in Kentucky, USA. A high-quality comprehensive central cancer registry is the basis for monitoring the cancer surveillance programme. It allows the identification of trends in incidence and stage of disease and guides planning and evaluation. The registry in Kentucky collects all data from 47 large and 62 small hospitals as well as pathology laboratories throughout the state concerning site-specific incidence, stage at diagnosis, histological type, treatment and outcome. This leads to the requirements for expansion of cancer screening programmes in some areas such as was demonstrated by the cervical cancer data from East Kentucky which identified a target population. The registry allows information from the stage at diagnosis as well as mortality data to assess the impact of the local screening procedures. This has been used in Kentucky to increase screening opportunities such as mammography in the areas where late-stage disease is predominating. The registry thus forms the model for cancer control, identifying the problem, selecting the target population and providing data on the impact of health policy. The reliance on a comprehensive data set to provide accurate feedback was emphasized.
|
SUMMARY OF SYMPOSIUM |
|---|
|
TOPINTRODUCTIONOPENINGSESSION 1SESSION 2: ORGAN-SPECIFIC...SESSION 3: PERSPECTIVES OF...SUMMARY OF SYMPOSIUMREFERENCES |
|---|
Chairperson: Dr R. Edward Hendrick
The symposium was closed with a summary given by Dr Matti Hakama (Tampere, Finland). The important issues raised by the symposium were highlighted with the balance between the detection of early treatable cancer and overdiagnosis. The summary followed the progress through the three days with papers examining the evaluation of the screening programmes and the seven primary organ-specific sites which were discussed with reference to the current situation in Japan and the rest of the world. Also discussed were the many new technologies which are emerging as potentially efficient and specific systems, the cost effectiveness requiring evaluation. It is evident that the main emphasis for the future will be on the development of new screening tests and genetic diagnosis may play some role in this. It is imperative that as new technology develops reasonable evaluation takes place in order to respond best to the benefits. Covered during the symposium were more than 30 different screening methods and it should be noted that although many are in widespread use, only three have any proven effect on mortality. The overall object of the screening programmes is to improve the health of the general population with tests that lead to a reduction in mortality, are safe and are cost effective.
|
Acknowledgements |
|---|
The organizing committee greatly appreciates the contributions of all the speakers and participants of this international symposium and their contribution to its ultimate success. This symposium was supported by the Foundation for Promotion of Cancer Research to promote the programme of the Second Term Comprehensive 10 Year Strategy for Cancer Control by the Ministry of Health and Welfare, Japan
|
FOOTNOTES |
|---|
+ For reprints and all correspondence: Tadao Kakizoe, Director, National Cancer Center Hospital, 1–1, Tsukiji 5-chome, Chuo-ku, Tokyo 104-0045, Japan
|
REFERENCES |
|---|
|
TOPINTRODUCTIONOPENINGSESSION 1SESSION 2: ORGAN-SPECIFIC...SESSION 3: PERSPECTIVES OF...SUMMARY OF SYMPOSIUMREFERENCES |
|---|
1 Sasako M, Mann GB, van de Velde CJH, Hirohashi S, Yoshida S. Report of the Eleventh International Symposium of the Foundation for Promotion of Cancer Research: Basic and Clinical Research in Gastric Cancer. Jpn J Clin Oncol 1998;28:443–9.
2 Abe K, Watanabe T. Report of the Twelfth International Symposium of the Foundation for Promotion of Cancer Research: Basic and Clinical Research in Breast Cancer. Jpn J Clin Oncol 1999;29:399–402.
3 Sugimura T. Multistep carcinogenesis: a 1992 perspective. Science 1992;258:603–7.
Received May 16, 2000; accepted May 17, 2000.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  F. Emura, D. Saito, T. Kakizoe, and M. J. Blaser Infection, Cancer and Prevention: Report of the 19th International Symposium of the Foundation for Promotion of Cancer Research Jpn. J. Clin. Oncol., November 1, 2006; 36(11): 745 - 755. [Full Text] [PDF]
|
|
|
|
|
|
P. S. Multani, K. Tobinai, T. Kakizoe, J. O. Armitage, R. Ohno, and T. Sugimura Report of the Fifteenth International Symposium of the Foundation for Promotion of Cancer Research: New Horizons in the Diagnosis and Treatment of Hematological Malignancies Based on Molecular Genetic Features Jpn. J. Clin. Oncol., September 1, 2002; 32(9): 371 - 385. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Collins, N. Shimoyama, E. Bruera, F. Takeda, T. Kakizoe, and T. Sugimura Report of the Fourteenth International Symposium of the Foundation for Promotion of Cancer Research: Pain Control, Palliative Medicine and Psycho-oncology Jpn. J. Clin. Oncol., September 1, 2001; 31(9): 459 - 468. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||