ies showing a reduction in deep melanomas or in mortality. Educational cam- paigns may have substantial effects on the demand for referral services [10,19]. In Queensland, a descriptive analysis by time period in one city looked at the effect of two public education campaigns, showing a 24% in- crease in the total number of melanocytic skin lesions excised in relationship to the campaigns [20]. The authors question the effectiveness of such cam- paigns, as they found no evidence for a change in thickness of melanoma or an increased detection rate. Free access skin checks These range from beach patrols and informal skin screening sessions in com- munity halls, to a process by which GPs or specialists will open their premises on occasions specifically to offer free skin checks. This approach has been used considerably in several countries, and large numbers of subjects have partici- pated [21–23]. Open access clinics have been assessed in the USA by the Amer- ican Academy of Dermatology (AAD) [24–27], in the Netherlands [28–31] and in Western Australia [32]. The participation rates in these clinics on a population basis are usually low, despite the large number of attenders. For example, the US total of some 282000 subjects screened in 1992–94 [27] translates to only about 2% of the at-risk population, if that is defined as white-skinned adults over the age of 30. This compares to participation rates of over 70% for many cervical smear and mammography screening pro- grammes. The detection rate of melanoma in that programme was 1.3 per 1000 screened, 90% being in situ or less than 0.76mm deep. The positivity rate is a central issue, both in its amount and its definition. Definingthis as the proportion of subjects given a label of suspected melanoma or something similar, gives modest positivity rates of between 1 and 4%, and the predictive value (the proportion of such patients who are subsequently found to have melanoma) is substantially high. The AAD programme, for ex- ample, defined 1.6% of participants as having suspected melanoma, and this was confirmed in 8.2% [27]. Even in the AAD programme, around 20% who screened positive apparently received no follow-up assessment or treatment. This may be acceptable as long as the service is seen as an extra voluntary effort, but is unlikely to be unacceptable for an established programme. Fail- ure to ensure follow-up in screening programmes for breast or cervical cancer has raised medicolegal issues, and screening for hypertension without follow- up has been shown to have detrimental effects [33]. Although the proportions identified as having suspect melanoma are mod- est, the proportions of subjects who have a positive screening result (in man- agement terms they are recommended to take further action) are much higher, being 10–12% in the Netherlands [28,31], 31% in the early Massachusetts PROSPECTS FOR POPULATION SCREENING 109 AAD programme [24] (reduced later) and 17% in Western Australia [32]. Any population screening programme with more than 10% of subjects being referred for further assessment creates a large requirement for follow-up services; in comparison, in breast and colorectal screening, the expected positivity rates would be closer to 5%. Most of the subjects referred were clin- ically labelled as having suspected basal cell carcinomas, squamous cell carci- nomas, or a range of other conditions including dysplastic naevi; all these conditions have such good outcomes in normal clinical practice that any im- provement from screening is unlikely. Melanoma screening programmes would be much more manageable and cost effective if only subjects with clinically suspected melanomas were referred. However, this requires clinical and medicolegal assurance that this is acceptable. If only those with clinically suspected melanomas are fol- lowed up, some melanomas will be missed; more selective referral accepts a re- duction in sensitivity for a substantial gain in specificity. In the Geraldton (Australia) survey [34], six out of 20 melanomas diagnosed were clinically labelled basal cell carcinoma, and would have been missed by a selective referral process. Examination of the entire skin has been recommended to avoid missing le- sions not noticed by the patient; in a US study, 13 of 14 melanomas found from an open access clinic were on body sites normally covered [35]. In the Nether- lands, in a voluntary self-referral clinic in which participants were asked to in- dicate whether they wanted one or a few specific lesions assessed or a complete skin check, there was a small increase in clinically suspicious lesions after the additional skin examination, and no change in confirmed melanomas. The authors question the value of offering full skin examination [36]. There are several studies of inter- and intraobserver agreement in skin can- cer or skin lesion assessment carried out after referral of patients to hospitals or clinics, but relatively few in a community setting. In one such study in the USA, kappa (k) values for interobserver agreement between dermatologists ranged from 0.38 for squamous cell cancer, based on only a very few cases, to 0.78 for squamous keratoses [37]. However, agreement on the simple man- agement issue of whether to refer or excise or not would be more valuable. The performance of nurses who had been given special training has been assessed [38,39]. In Western Australia [39], nurses showed a sensitivity of 95% and a specificity of 84% compared to a surgeon’s clinical assessment, showing that they could be more cost effective. Nurses as screeners have also been assessed in Sweden [40] where nurses referred some 10% of subjects as having suspi- cious lesions, compared to 3.5% referred for biopsy by doctors. The benefits of open access clinics seem dubious, although they may be valuable for their stimulation and educational effects on both the profession and the public, and they may lead to more systematic programmes being de- veloped later. 110 CHAPTER 9 Case finding at primary care level This means questioning a patient who visits their primary care provider for an- other reason about recent skin changes, or offering a partial or full skin exam- ination. This can be promoted both by encouraging the general public to ask for it and encouraging GPs to offer it. Any general education or self-screening campaign will also tend to encourage this behaviour. Improved case finding, allied to improved referral decisions, and prompt and accurate management at the referral level, is probably the most important management tool to re- duce melanoma mortality. Many countries have documented considerable improvements in survival over time, together with improvements in depth distribution and diagnosis, and it seems likely that this is a result of a combi- nation of increased public awareness of the early signs, better management by primary care doctors, and better and faster hospital referrals and treatment. However, documentary evidence from controlled studies is almost non- existent, and there are few specific studies of the effects of efforts to improve case finding at primary care level. There is evidence in high-risk countries that the level of expertise in GPs is high. In a survey of a large representative sample of GPs in New Zealand, over 95% described the correct management for three presented scenarios involv- ing early melanoma, and the responses were not greatly different from those of a small sample of specialist dermatologists [41]. Skill levels were higher in younger than in older GPs, and were increased in those who had dealt person- ally with at least one melanoma patient. In Australia, the effect of a one-day training course for GPs on their diagnostic abilities was tested by examination of patients at clinics set up with a higher than normal proportion of subjects with suspicious lesions [42]. The training had no significant effect on the sen- sitivity, specificity or predictive value of the screening. In general, the sensitiv- ity was high but specificity low: many subjects with non-malignant lesions were referred for further assessment. The scope for further improvements in case finding and primary care may be limited in high-risk areas unless screening activities are added. However, it seems likely that there is considerable potential for improvement in moderate- and low-risk countries where the level of public awareness, the expertise in primary care management, and the speed and appropriateness of referral practices may all have room for improvement. In the UK, delays both in the presentation of suspicious lesions to GPs and in the referral process have been documented, and setting up specific pigmented lesion clinics may have some success [10,43,44]. Self-screening This goes considerably further than public education, although obviously PROSPECTS FOR POPULATION SCREENING 111 there is an overlap. While any campaign which attempts to educate about the early signs of melanoma must encourage people to be aware or look at their skin, some initiatives specifically encourage a specific self-screening protocol. This can range from a simple checklist [45] to the comprehensive regular screening promoted by the American Cancer Society [46]. Self-screening has been assessed in a case–control study in Connecticut [47]. This study compared subjects with ‘lethal’ (fatal or advanced) melanoma with population controls, assessing the history of self-examination, defined as ‘a careful, deliberate and purposeful examination of the skin’, by question- naire. The results showed a substantial but non-significant reduction in the risk of ‘lethal’ melanoma among melanoma patients (risk ratio 0.58, 95% CI = 0.31-1.11), which is consistent with a beneficial effect. However, the study also showed a reduction in total melanoma incidence associated with screen- ing (risk ratio 0.66, 95% CI = 0.44-0.99), which is counterintuitive. The main way in which incidence could be reduced is by the recognition and removal of precursor lesions, which seems an unlikely effect and there is no direct evi- dence of this from the study [48]. By assuming both effects are valid, the au- thors estimated that self-examination may reduce mortality from melanoma by 63% (risk ratio 0.37, 95% CI = 0.16-0.84). However, the reduction in in- cidence suggests bias or confounding with the study, raising questions about the validity of the other results. While this is a carefully performed study, as- sessing screening by case–control methods is inherently difficult, and further assessment of screening by analytical studies or preferably by randomized trials is required. Recent Australian data show a high level of screening. In New South Wales, 48% of subjects reported self-screening and 17% reported having had a GP skin check in the previous year [49]. However, screening involving a sys- tematic whole body assessment is much less common. In a Queensland survey [50,51], 60% of subjects reported that they practised self-screening, but the authors felt that many or most of the examinations would be inadequate. In a telephone survey of a national sample in the USA [52], almost half the respon- dents reported that they conducted skin self-examination and, of those that did, the majority reported that they examined their skin at least weekly or even daily. It is questionable whether weekly or daily screening is a healthy behav- iour, but the thoroughness of the self-examination was not described. In the USA, performance of self-screening (ever) was positively associated with self- perceived risk and with having discussed skin cancer with health professionals [53]. Thus there is some evidence for benefit from self-screening and further as- sessment is warranted. Evaluating the practice of self-screening is difficult be- cause it is under individual control; the vigorous promotion of self-screening on an individual or population basis could be evaluated in a trial. 112 CHAPTER 9 Invitation-based screening by the primary health care provider The term ‘screening’ can be confusing [54]. The essential difference between screening and case finding is that case finding is opportunistic, adding an extra question or examination, but only for those people visiting the GP for other reasons. In contrast, screening, as it is used in other disease contexts, implies a deliberate visit to the GP (or other screener) for that purpose. There will need to be a specific programme to encourage people to come for screening. Thus, a GP practising case finding would offer skin examination to all patients who visit his or her premises, while a GP practising skin screening would use the age–sex register to invite all patients in the practice (perhaps over a certain age) for screening at regular intervals. Skin screening is therefore a new pro- gramme in addition to normal care, with financial and organizational impli- cations. A specific invitation-based screening service by the primary care provider has considerable potential and deserves fuller assessment. No trials of such a programme have been published. New screening services outside normal care provision A new screening service outside the normal care provision mechanism could consist of a screening clinic, allied to a programme to ensure that all residents in the area were made aware of the new facility and encouraged to attend. This could involve media publicity or a personal invitation, analogous to the meth- ods used in breast or cervical screening. Such a system apparently has yet to be tried in any large general population in regard to melanoma. Screening has been used in England in a privately funded (BUPA) general health screening centre [55]. A complete skin check was carried out by a doctor, usually a GP, and the position and characteristics of any pigmented lesion regarded as sus- picious or changing were marked on a skin chart, and assessed by the seven- point checklist system. All lesions were then photographed using Polaroid cameras, and the picture assessed independently by two consultant dermatol- ogists. Of 39922 subjects screened, 948 (2.4%) had at least one skin lesion as- sessed and photographed. Of the 1052 lesions in these subjects, 231 were assessed as requiring follow-up and amongst these there were 11 melanomas, but follow-up was incomplete. The authors conclude that photography great- ly reduced the need for specialist referral, but their estimate of sensitivity was only 37%. Selective screening of high-risk groups All the above have dealt with general population approaches, which may be restricted by age but not in other ways. Other approaches are designed to tar- PROSPECTS FOR POPULATION SCREENING 113 get high-risk subgroups. The impact on the whole disease depends on both the impact on the selected subgroup, and also how that subgroup is selected and what proportion of all melanomas in the population occur within the group. Most of the above methods, particularly the free access clinics, have an ele- ment of high-risk selection, although this is unsystematic and subjective. Sys- tematic attempts to identify high-risk groups on a population basis are more complex, as they require a preselection process for risk which is itself a form of screening. The highest risk groups are subjects with a previous melanoma, the dys- plastic naevus syndrome, or specific genetic markers; these are dealt with else- where in this book. Careful regular surveillance, based on clinical observation supplemented by photography, has been shown to result in improved depth distribution of subsequent melanomas [56–58]. There are no randomized tri- als of alternative types of follow-up surveillance in high-risk groups. The value of screening for less well-defined risk groups, such as subjects with some dys- plastic naevi or a high total naevus count, is less clear [59–61]. Limitations of screening A major limitation of melanoma screening is that it is simply using unaided observation. Most current efforts are based on the clinical signs and symp- toms of melanoma, which are non-specific and have a low predictive value. In New South Wales, a population survey showed that 12% of adults had had within the previous 12 months a lesion fitting the ABCD criteria for a suspicious lesion [49]. A study of 21-year-olds in New Zealand showed a 20% frequency of the same phenomenon, with approximately half of these subjects having sought a primary care opinion [62]. In a general practice in England, self counting of naevi was compared with a naevus count carried out by trained GPs [63]. The authors report that self-counts could be useful in self- identification of risk, and note that in this group 13% had reported some change in a mole in the previous 3 months and 5% had reported a major change. Instruments such as the dermatoscope, and techniques of computer image analysis [64–68] are likely to be valuable, mainly in regard to the diag- nosis of referred patients and in the regular follow-up of high-risk subjects with specific lesions. The application of these methods for population screen- ing requires much further development and evaluation in a population con- text. Simpler technology may be valuable; a randomized trial in Queensland showed that providing GPs with a simple algorithm and a cheap camera to as- sist their follow-up of patients with suspicious skin lesions gave a reduction in excision rates of benign lesions, with no change in the excision rates of melanoma [69]. 114 CHAPTER 9 Number needed to screen per death prevented A very approximate estimate can be produced from simple calculations. Con- sider a high-risk general population, Australia, and the age range 50–69, in both sexes, as a feasible target for screening. Their annual mortality rate from melanoma (1996) is 10 per 100000. Suppose, optimistically, a 30% reduction in mortality from a screening test every 2 years for which the positivity rate (the proportion of screenees requiring further investigation) is 5% on the first screen, reducing to 2% on the second screen. Assume that the cost per screen is $A24, and the cost for the investigation of each positive subject is $A140 [70]. On this basis, to prevent one death would require screening 1700 persons (number needed to screen; NNS) and performing some 18000 screens. The cost would be around $A500000 and the number of false-positives to be in- vestigated would be somewhat over 400. If we assume a life saved equates to 20 years of lives saved, the cost per year of life saved would be about $A25000 (approximately £10000), which is comparable to the costs of breast or colo- rectal cancer screening. As the mortality rate is higher in men, the NNS is around 1200 for men and 2600 for women in this age group. However, a full assessment is much more complex, as it needs to assess the marginal costs and benefits of screening compared with current practice, which includes a high level of less systematic assessment, many GP visits, referrals and biopsies, and needs to use appropriate discounting. A cost effectiveness analysis has been published [70] for the Australian sit- uation which assumes mortality reductions of 15–34% with either 5- or 2- yearly screening, starting at age 50. This produces costs per year of life saved, with appropriate analytical assumptions including discounting, of from $A6800 (men: 5-yearly, 60% sensitivity, 27% mortality reduction) to $A31000 (women: 2-yearly, 30% sensitivity, 24% mortality reduction). These costs are comparable to those of breast cancer screening. The critical as- sumption is that screening does produce a substantial reduction in mortality. Further, generally similar, estimates have also been given [42]. One of the main contributions to the cost is the fact that the death rate from melanoma, even in high-risk countries, is relatively small compared, for ex- ample, to the death rate from breast cancer or colorectal cancer. If the above simple calculation is based on UK annual mortality rates for subjects aged 50–69 (approximately 3.5 per 100000), the number needed to be screened to prevent one death is about 4800, with 48000 screens. Positivity rates should be lower; using 2% for a first screen and 1% thereafter gives about 500 posi- tives to be assessed per death prevented, and a cost per year of life saved of around $A60000 (£24000). Obviously, the results of these calculations will be more favourable if the mortality rate is increased, as would be possible if a high-risk group is defined, but it is then important to add in the costs, and the PROSPECTS FOR POPULATION SCREENING 115 accuracy or otherwise, of whatever method is used to define the high-risk group. The evidence base for melanoma screening Screening for cancer on a population basis requires strong evidence. Screening for breast cancer and colorectal cancer is supported by evidence from large- scale population-based randomized trials; screening for uterine cervical can- cer does not have randomized trial evidence, but has a large number of strong cohort and case–control studies to support it. In other cancer screening situa- tions, for example ovarian cancer, the predominant view is that screening should not be undertaken until the results of current randomized trials be- come available. In contrast, the evidence base for screening for melanoma is extremely weak (Table 9.2). There are no available results from randomized trials or cohort studies, and the only analytical study result being the single case–control study. The main argument for the effectiveness of screening is based on the assumption that earlier diagnosis will produce mortality and morbidity benefits, which in turn is based on the large differences in postdiag- nosis survival with depth of invasion, for patients diagnosed in normal clinical practice. These assumptions may be correct, but the lack of empirical evidence of benefit needs to be acknowledged. Conclusions In high-risk and high awareness countries, such as Australia and New Zealand, the further potential for media education on the early signs of 116 CHAPTER 9 Table 9.2 Levels of evidence (from the Physician Data Query of the US National Cancer Institute) applied to cancer screening Level of evidence Cancer screening application 1 Evidence obtained from at least one RCT Breast, colorectal (several RCTs) 2 Evidence obtained from controlled trials that Cervix (many studies) use allocation methods other than randomization (e.g. by birth date or hospital chart number) 3 Evidence obtained from cohort or case–control Melanoma (one case–control study) analytic studies, preferably from more than one centre or research group 4 Evidence obtained from multiple time series with or without intervention (quasi-experimental designs) 5 Opinions of respected authorities based on clinical experience, descriptive studies, or reports of expert committee Abbreviation: RCT, randomized controlled trial. melanoma is probably limited and open access clinics are of dubious value. Case finding in primary care, while it may be quite good at present, has con- siderable potential for further development, and innovative programmes could be useful. The promotion of self-screening may be of value; assessing costs and outcomes is a priority. Invitation-based screening in primary care should be developed on a pilot basis and assessed. Invitation-based screening as a new service is probably not very valuable in countries with a well- developed primary care service, if assessed in terms of skin screening alone. The potential for an integrated invitation-based screening service for a number of chronic diseases is greater. In lower risk countries with lower levels of public and professional aware- ness, such as the UK, improved public and professional education has been shown to be valuable, and programmes including spot checks and specific pro- fessional training may be useful. While the frequency of the disease is lower, the relatively low levels of awareness increases the potential benefits of educa- tion and screening programmes. Further development should pay great attention to evaluation, using the most rigorous methods which are feasible. The best method of assessment is a randomized trial, and a large randomized trial of a community-based pro- gramme using education, self-screening and doctor screening is being devel- oped in Queensland (J. Aitken, personal communication). Other randomized trials would be valuable. Carefully performed evaluations using cohort, case–control, or time series designs will be valuable if control groups are used, and attention is paid to bias and confounding. References PROSPECTS FOR POPULATION SCREENING 117 1 Rigel DS, Friedman RJ, Kopf AW. The incidence of malignant melanoma in the United States: issues as we approach the 21st century. J Am Acad Dermatol 1996; 34: 839–47. 2 Giles GG, Armstrong BK, Burton RC, Staples MP, Thursfield VJ. Has mortality from melanoma stopped rising in Australia? Analysis of trends between 1931 and 1994. Br Med J 1996; 312: 1121–5. 3 Roush GC, McKay L, Holford TR. A reversal of the long-term increase in deaths attributable to malignant melanoma. Cancer 1992; 69: 1714–20. 4 Roder DM, Luke CG, McCaul KA, Esterman AJ. Trends in prognostic factors of melanoma in South Australia, 1981–92: implications for health promotion. Med J Aust 1995; 162: 25–9. 5 Balch CM, Soong SJ, Shaw HM, Urist MM, McCarthy WH. An analysis of prognostic factors in 8500 patients with cutaneous melanoma. In: Balch CM, Houghton AN, Milton GW, Sober AJ, Soong SJ, eds. Cutaneous Melanoma. Philadelphia: J.B.Lippincott, 1992: 165–87. 6 Häffner AC, Garbe C, Burg G, Büttner P, Orfanos CE, Rassner G. The prognosis of primary and metastasising melanoma: an evaluation of the TNM classification in 2495 patients. Br J Cancer 1992; 66: 856–61. 7 Burton RC, Coates MS, Hersey P, et al. An analysis of a melanoma epidemic. Int J Cancer 1993; 55: 765–70. 8 Burton RC, .Armstrong BK. Recent incidence trends imply a non- metastasizing form of invasive melanoma. Melanoma Res 1994; 4: 107–13. 9 Burton RC. Analysis of public education and the implications with regard to nonprogressive thin melanomas. Curr Opin Oncol 1995; 7: 170–4. 10 Whitehead SM, Wroughton MA, Elwood JM, Davison J, Stewart M. Effects of a health education campaign for the earlier diagnosis of melanoma. Br J Cancer 1989; 60: 421–5. 11 Doherty VR, MacKie RM. Experience of a public education programme on early detection of cutaneous malignant melanoma. Br Med J 1988; 297: 388–91. 12 MacKie RM, Hole D. Audit of public education campaign to encourage earlier detection of malignant melanoma. Br Med J 1992; 304: 1012–15. 13 MacKie RM. Strategies to reduce mortality from cutaneous malignant melanoma. Arch Dermatol Res 1994; 287: 13–15. 14 MacKie RM. Melanoma prevention and early detection. Br Med Bull 1995; 51: 570–83. 15 Herd RM, Cooper EJ, Hunter JA, et al. Cutaneous malignant melanoma: publicity, screening clinics and survival — the Edinburgh experience 1982–90. Br J Dermatol 1995; 132: 563–70. 16 Melia J. Early detection of cutaneous malignant melanoma in Britain. Int J Epidemiol 1995; 24: S39–S44. 17 Bonerandi JJ, Grob JJ, Cnudde N, Enel P, Gouvernet J. Campaign of early detection of melanoma in the Provence–Alpes– Cote-d’Azur area 1989: lessons of an experience. Ann Dermatol Venereol 1992; 119: 105–9. 18 Bulliard J-L, Raymond L, Levi F, et al. Prevention of cutaneous melanoma: an epidemiological evaluation of the Swiss campaign. Rev Epidemiol Sante Publique 1992; 40: 431–8. 19 Graham-Brown RAC, Osborne JE, London SP, et al. The initial effects on workload and outcome of a public education campaign on early diagnosis and treatment of malignant melanoma in Leicestershire. Br J Dermatol 1990; 122: 53–9. 20 Del Mar CB, Green AC, Battistutta D. Do public media campaigns designed to increase skin cancer awareness result in increased skin excision rates? Aust N Z J Public Health 1997; 21: 751–3. 21 Marks R. Two decades of the public 118 CHAPTER 9 health approach to skin cancer control in Australia: why, how and where are we now? Australas J Dermatol 1999; 40: 1–5. 22 McCarthy WH. Secondary prevention of skin cancer in Australia. In: MacKie RM, ed. Primary and Secondary Prevention of Malignant Melanoma. Basel: Karger, 1996: 31–42. 23 Howell JB, Cockerell CJ. Melanoma self- examination day: Melanoma Monday, May 1, 1995. J Am Acad Dermatol 1996; 34: 837–8. 24 Koh HK, Caruso A, Gage I, et al. Evaluation of melanoma/skin cancer screening in Massachusetts. Cancer 1990; 65: 375–9. 25 Koh HK, Geller AC, Miller DR, Caruso A, Gage I, Lew RA. Who is being screened for melanoma/skin cancer? Characteristics of persons screened in Massachusetts. J Am Acad Dermatol 1991; 24: 271–7. 26 Koh HK. Geller AC. Melanoma control in the United States: current status. [Review] Recent Results Cancer Res 1995; 139: 215–24. 27 Koh HK, Norton LA, Geller AC, et al. Evaluation of the American Academy of Dermatology’s national skin cancer early detection and screening program. J Am Acad Dermatol 1996; 34: 971–8. 28 Rampen FHJ, van Huystee BEWL, Kiemeney LALM. Melanoma/skin cancer screening clinics: experiences in the Netherlands. J Am Acad Dermatol 1991; 25: 776–7. 29 Rampen FH, Casparie-van Velsen JI, van Huystee BE, Kiemeney LA, Schouten LJ. False-negative findings in skin cancer and melanoma screening. J Am Acad Dermatol 1995; 33: 59–63. 30 de Rooij MJ, Rampen FH, Schouten LJ, Neumann HA. Skin cancer screening focusing on melanoma yields more selective attendance. Arch Dermatol 1995; 131: 422–5. 31 de Rooij MJ, Rampen FH, Schouten LJ, Neumann HA. Volunteer melanoma screenings: follow-up, compliance, and outcome. Dermatol Surg 1997; 23: 197–201. 32 Katris P, Crock JG, Gray BN. Research note: the Lions Cancer Institute and the Western Australian Society of Plastic Surgeons skin cancer screening [...]... Townsend PL A pigmented lesion clinic: 45 46 47 48 49 50 51 52 53 54 55 119 analysis of the first year’s 1 055 patients Br J Plast Surg 19 95; 48: 247 51 Fitzpatrick TB, Rhodes AR, Sober AJ, Mihm CM Jr Primary malignant melanoma of the skin: the call for action to identify persons at risk; to discover precursor lesions; to detect early melanomas In: Elwood JM, ed Naevi and Melanoma: Incidence, Interrelationships... physicians: a cost effectiveness analysis J Med Screen 1996; 3: 47 53 Melanoma: Critical Debates Edited by Julia A Newton Bishop, Martin Gore Copyright © 2002 Blackwell Science Ltd Part 3: Management Melanoma: Critical Debates Edited by Julia A Newton Bishop, Martin Gore Copyright © 2002 Blackwell Science Ltd 10: Excision of primary cutaneous melanoma Michael J Timmons Introduction Complete surgical excision... Surg 1986; 73: 793 5 McLean NR, Boorman JG Secondary malignant melanoma arising in a contralateral thigh donor site Br J Plast Surg 1984; 37: 386–7 Rigby HS, Briggs JC Subungual melanoma: a clinico-pathological study of 24 cases Br J Plast Surg 1992; 45: 2 75 8 Park KGM, Blessing K, Kernohan NM Surgical aspects of subungual malignant melanomas Ann Surg 1992; 216: 692 5 Heaton KN, El-Naggar A, Ensign... subungual melanoma Ann Surg 1994; 219: 197–204 33 Quinn MJ, Thompson JE, Crotty K, McCarthy WH, Coates AS Subungual melanoma of the hand J Hand Surg 1996; 21A: 50 6–11 34 Ross MI, Balch CM Surgical treatment of primary melanoma In: Balch CM, Houghton AN, Sober AJ, Soong S-j, eds Cutaneous Melanoma, 3rd edn St.Louis: Quality Medical Publishing, 1998: 141 53 35 Hayes IM, Thompson JF, Quinn MJ Malignant melanoma. .. cm are safe for melanomas < 0.76 mm thick [36,37] Although there have been no published long-term follow-up studies of such narrow margins, there is a case for excision of minimally invasive small melanomas with a total margin of 3 5 mm of normal skin, treating them similarly to melanoma in situ However, the WHO trial of 1 cm and 3 cm margins included 112 patients with melanomas 0.1– 0 .5 mm thick and... Surg 1984; 20: 759 –63 17 Milton GW, Shaw HM, McCarthy WH Resection margins for melanoma Aust NZ J Surg 19 85; 55 : 2 25 6 18 Griffiths RW, Briggs JC Incidence of locally metastatic (‘recurrent’) cutaneous malignant melanoma following conventional wide margin excisional surgery for invasive clinical stage I tumours: importance of maximal primary tumour thickness Br J Surg 1986; 73: 349 53 19 Roses DF, Harris... Oncol 1996; 3: 446 52 Heaton KM, Sussman JJ, Gershenwald JE, et al Surgical margins and prognostic factors in patients with thick (>4 mm) primary melanoma Ann Surg Oncol 1998; 5: 322–8 Briggs JC, Ibrahim NBN, Hastings AG, Griffiths RW Experience of thin cutaneous melanomas ( . depth distribution of subsequent melanomas [56 58 ]. There are no randomized tri- als of alternative types of follow-up surveillance in high-risk groups. The value of screening for less well-defined risk groups,. 19 95; 51 : 57 0–83. 15 Herd RM, Cooper EJ, Hunter JA, et al. Cutaneous malignant melanoma: publicity, screening clinics and survival — the Edinburgh experience 1982–90. Br J Dermatol 19 95; 132: 56 3–70. 16. been published [70] for the Australian sit- uation which assumes mortality reductions of 15 34% with either 5- or 2- yearly screening, starting at age 50 . This produces costs per year of life