Prevention and control of melanoma: the public health approach

Where they have been recorded, incidence and mortality rates for melanoma have been rising steadily (although with some variation between countries) during the 20th century. Although there may be some breakthroughs with novel treatment on the horizon (as highlighted in this issue), at present surgical excision is the mainstay for cure of this disease. There is little doubt that detection and excision of the tumor in an early stage have contributed most to the decreasing case fatality rates seen in recent decades. Early detection of melanoma is an approach to control that, if used effectively, can have a relatively rapid impact on decreasing population mortality rates. It focuses on people who either have melanoma or will develop it in the very near future. It is the first approach taken by most countries when implementing a public health control program for melanoma. In the presence of a continuing increase in the incidence rate of melanoma, another approach considered by many cancer control programs is primary prevention. In this case, a cause for the disease (sunlight exposure) is highlighted, and programs are developed to reduce exposure to this environmental risk factor. It is a long-term approach. Because there is a considerable time lag from exposure to sufficient sunlight to induce and then promote melanoma and the onset of the clinically apparent tumor, it is likely that decades must pass before a detectable reduction in the onset of new tumors occurs as a result of reduced exposure to sunlight following a melanoma primary prevention program. Nevertheless, primary prevention and early detection (secondary prevention) are the two major public health approaches now being considered or undertaken on a population basis in many countries. This article considers both approaches. It outlines briefly the data and assumptions on which the approaches are based and gives some examples of program development, delivery, and measurement of outcome. Epidemiologic and behavioral research underpin the process in both primary prevention and early detection programs. They provide data on the need, the cause, the susceptible population, the targets for change, and assessment of effect of the public health approach to melanoma control. To influence mortality due to a cancer, the tumor must be detectable in its primary tissue at a stage when it can be easily treated before secondary dissemination is likely to be lethal within the affected individual's expected lifetime. There are data now available which show that survival from invasive cutaneous melanoma can be influenced by excision when the tumor is thin. Using Breslow's microscopic measurements (the thickness of the melanoma in millimeters from the granular layer of the epidermis to the deepest portion vertically below), it is possible to predict the likelihood of metastatic disease and subsequent mortality.1 Combined with other prognostic variables (e.g., sex, site, ulceration, Clark's level of invasion, and others), a reasonably accurate risk of metastasis can be calculated.2 Using thickness alone, it has been shown that tumors less than 0.75 mm in thickness are likely to be lethal in less than five percent of patients. Tumors that are 4 mm or more in thickness at primary excision have developed the capacity to metastasize and be lethal in more than 50 percent of patients.2 The success of the public health approach to early detection relies on an ability to detect thin, early tumors. This in turn requires that early melanoma has characteristics that highlight the fact that it is abnormal. These characteristics must develop at a rate that allows them to be noticed and treated before the tumor has developed sufficiently to be potentially lethal. Thus, there is a need to have the melanoma seen (the public) and removed (the health professional). There are very few data available that give a clear indication of the natural history of clinically apparent melanomas. Nevertheless, anecdotal data suggest that many tumors grow relatively slowly over months to a year or more in a thin stage (lateral growth phase), before becoming deeply invasive (the vertical growth phase).3 These tumors include the superficial spreading melanomas and the lentigo maligna melanomas. Lentigo maligna melanomas may grow very slowly for many years on sunlight-exposed areas, such as the face, before becoming deeply invasive. Both lentigo maligna melanoma and superficial spreading melanomas develop clinical characteristics that suggest a need for attention. These can be remembered by thinking of ABCD: A (asymmetry); B (border irregularity); C (color variegation); and D (diameter enlarging) (Table 1).4 More recently, MacKie5 has published a seven-point checklist for medical practitioners refining the clinical characteristics to major signs and minor signs. A study of the original seven points suggested that of the various signs, border irregularity is the most powerful predictor of a melanoma.6 However, this simple sign still only had a positive predictive value of 13 percent. Superficial spreading and lentigo maligna melanomas are most suitable for the public health approach to melanoma control. Nodular melanoma, on the other hand, appears as a rapidly growing tumor in the vertical growth phase from an early stage. Nodular melanomas may not show all the typical characteristics of melanoma described above and thus are difficult tumors to detect in an early curable phase. They make up 10 to 20 percent of melanomas treated in most series reported. Unfortunately, they tend to have a higher case fatality rate due to detection at a thicker stage than superficial spreading melanoma or lentigo maligna melanoma. Therefore, they contribute proportionally more to mortality data than the other types. In designing early detection programs, it is desirable to be able to concentrate on populations at high risk of developing the tumors (Table 2). From the data on risk factors for melanoma, it is clear that the most important risk factor is skin color. Melanoma is very rare in black people. Among people who are not black, the major risk factor is number of melanocytic nevi (moles), both common acquired and dysplastic or atypical.7 Risk gradients have been calculated for numbers of nevi, with greater than 200 nevi considered high risk (relative risk greater then 20 in some series). Other risk factors include tendency to freckle, unprotected skin response to strong sunlight (burn/tan response), family history of melanoma, family history of atypical nevi, past history of excessive sunlight (burns), and country of childhood.8 As in most cancers, age itself is a risk factor. It is difficult epidemiologically to separate age from exposure to sunlight. People with a family history of melanoma and dysplastic nevi who have a large number of nevi themselves are at very high risk of developing a melanoma over a lifetime.9 In some series this cumulative lifetime risk approaches 100 percent. Thus, there is a case for placing such individuals under surveillance and regular follow-up. The use of clinical photography to record baseline lesions and detect change over time is a useful technique. Early tumors have been detected in follow-up of these patients in specialized clinics. Interestingly, it is the clinicians using the photographs who have detected the vast majority of melanomas, rather than the patients themselves.10 Apart from this very high-risk group, the role of screening as a population-based approach to melanoma control is controversial. Although the risk factors mentioned above do help pinpoint which particular people are at increased risk, in most populations they account for only around 30 to 50 percent of people who develop lethal tumors. As yet, there are no data on population-based mass screening of asymptomatic people demonstrating a reduction in mortality rate. Poster used in public education programs in Australia. It highlights the other, more common skin cancers as well as melanoma. Reproduced with permission from the Anti-Cancer Council of Victoria. On the other hand, a reduction in mortality rate has been reported in some populations following public education programs.11,12 These programs have relied on members of the public inspecting their own skin (or that of a friend or partner) and then visiting a health for subsequent This is primary screening by skin There have been a of for skin as there have been no that with the of particular A recent by that skin may be of some this is very early and as the themselves more research is before are to increase the of particular technique. There are no to suggest the with which skin be There are data published that members of the public do their own A study of people in that percent had noticed or more signs of on the in the The study that the positive predictive value of the is around percent. study that members of the public may be able to if they are at high risk using and as risk A of patients melanoma able to show a of percent for number of on the percent for or more nevi on the and percent for detection of nevi than mm in on of the from to 100 on the of and number being the the signs to for has been the mainstay of education programs that have of people to a health for relies on the ability of the to be able to and to the to with of public health program that relies on to provide and of detected by the is to to the public education program. In a skin cancer skin cancer reported to have an of of and positive predictive value of percent for detection of melanoma in the years to In for skin cancer by the of positive predictive value for by to These are data for not in melanoma have rates than Using photographs as a study that only percent of or more of melanomas with percent of They that only percent of dysplastic nevi with percent of There are a of the in of of melanoma and other lesions between and less well In some specialized by in melanoma have been In other members of the public can from a such as a as the first of This is the case in the in the of health may to the primary being first in the future. This is the case in there are very few lesions In these education programs need to be at the of first from most countries with relatively high mortality rates for melanoma are that it to in the at a higher rate than people. This is the case in years or In the lethal tumors with on of the patients medical practitioners for other medical (e.g., of or There may be the are not early for a tumor that be These may include many other skin lesions it difficult to detect an early other health and a of a to a change or the to medical attention. this there is a case for on primary and screening their patients. This is the to the skin of patients when they present for another or a of There are data that 20 percent or more of melanomas in some series are detected by a rather than a There are the in of melanoma between and The use of skin cancer with for as occurs in the as a of the of and the provide an to the public to be of and from a health On the other hand, it is to that these provide a of screening for melanoma on a population Although large numbers of the public have been seen and many melanomas there has been no of the impact of these on long-term or mortality from melanoma in the There may be a case for such as to as in populations of melanoma is likely to be As this has not been to screening and detection for melanoma at present relies on In an to the the has been and used in a number of This is an which on the skin to reduce the and of the epidermis and The use of this is still in an early stage. when this used by who not in its to than when in its with exposure to many people with melanoma or other its value is at role may well in of thus rather than in detection of melanoma. of lesions is under study in a number of There have been in the of this approach in a exposure to Nevertheless, as and there may be a case for such an in early detection of This is likely to be when using the for detection of change within the over The of early detection of melanoma is to reduce the mortality rate due to the A secondary of early detection is a reduction in the and all its of programs can account of of the the from by the public for assessment and then of tumors to subsequent and mortality to melanoma (Table The first phase of early detection programs education and of the population and the health to to in medical and following education programs be a of change in for following an education program be a of effect of a program the As there are in the likelihood of assessment of change in and melanoma and its treatment be an important of an at the A of be to not only of the of a melanoma The thickness of tumors removed before and education programs is a of published in the increase in the of tumors removed in the very thin is as of a However, there have been that education programs can result in detection of tumors that may not have to a lethal Thus, their detection may the assessment of It is a reduction in mortality due to very tumors that is the primary of these programs. Thus, it is the numbers of these tumors that are detected on a continuing basis that the success in the A reduction in numbers of tumors the for a population over time is the In the there may be an in the numbers of people with tumors detected as a result of the following there be a as more people have their tumors detected early and less have them develop and lethal A population-based tumor is important in the effect of a program on the the program only people with early tumors to present for the will show that there has been no effect on the numbers of people still with tumors. In many the to all tumors lethal is a to of a melanoma control program. a reduction in mortality rate is the of Because of time with of thin it is to use long-term survival rates such as survival data as the of rather than survival and of these data is to the and subsequent of these programs. of early detection programs may be in a number of of a of or time and of and subsequent prevention of melanoma a of the a large number of possible factors being to exposure to sunlight by people at risk of melanoma is the only environmental factor to melanoma in most The risk factors (or mentioned in the The of sunlight exposure to risk of melanoma is not Epidemiologic suggest it is high sufficient to cause in that are There are in risk between and and within These are on which suggest that it is not only exposure that is Nevertheless, that high exposure to sunlight during the for high rates of melanoma as an of as a is a risk it is not clear this is a indication of very large at that age or a of skin to excessive sunlight. demonstrating the of a a a Reproduced with permission from the Anti-Cancer Council of Victoria. There is no for the of sunlight that cause melanoma. It has been to be in the to of of this ability to cause More it has been suggested from on of that the to may contribute as Although these are it be that it is exposure to the of sunlight that has the high incidence rate Thus, be from the of sunlight are a little in of the of the various of sunlight to the of this There may be in that reduce only a of some large to has a new to the that many people the need for primary prevention programs. have been of increased incidence and mortality rates as a result of increased by In they have to be rather and may well have the likely outcome. most people that environmental be of the approach to primary prevention is to not only change that to excessive to an increase in the or of the sunlight to which are As yet, have no very of individuals likely to develop melanoma if to sufficient sunlight in the at the screening is becoming may a of the population to primary prevention programs be skin populations who or in are the primary the of data on a for melanoma in the public health the simple of an reduction in exposure to sunlight. this concentrate on people to reduce the risk of when data suggest that there are some during which the risk of is higher than have the of people when at This be due to the of from and the effect of which a of A effect is seen on reduce the of sunlight and thus reduce its On the other hand, they may have little or no effect on the common that the following exposure on a is rather than Risk of during well with There is a more of to are and are are are In people up or to In people are more likely to of In they are most likely to research data show a clear of risk of with for a Thus, an for education programs be to the and the of a Although there has been some on this can be in primary prevention programs have been on the there have been in and for of the to which this has has been in a in the use of for not only in to that they rather than in with age and It is higher in and than and higher in than approaches to with age and are more likely to be used by than and by people than people. On the other hand, are more likely to be used by than and by people than of the points mentioned is from behavioral They are useful in targets and points for developing subsequent education programs. A to be used with primary Reproduced with permission from the Anti-Cancer Council of Victoria. There are a number of in which sunlight exposure can be reduced from of around the of the and and can all to in is as the be as an to natural not a for A at percent of is in the two either of to or to during by and other can reduce The higher the the the more there is for to the people under the with being high for and Although still may be when the is vertically above to 10 is This can to high of incidence for the and increase This is when from and are with can provide A of 10 can to a reduction of percent in exposure to the and The of of is more important than the a to Because it may a that allows to be is to make it the and of the can by more and thus less than which a a may more There have been in a number of countries to for and the of A simple of is to the up to strong and it a or it is to A recent article on that a large of available not for had a of around 20 or are for their under A factor is by the increase in the of to cause cutaneous in using the with the when the skin is a the of is there is developing a in of in the the reduction of with It is an response with relatively little reduction in for relatively in a level of around In most health use of a that is to some and that is at with a higher are is This that with their relatively role in it is to the increased and other to regular of very high of people in to more people are at from The is a of under There are in it to a effect in use (e.g., for individual's time on particular is on many These include site, of skin of thickness of the epidermis time of time of and from and It to an individual's time at particular and thus using numbers to time before is the number of and other to exposure are to be for the that it is to promote to reduce risk from to sunlight. In that using regular rather than up or on a basis to some it be to suggest that people use in a only when they there will be a of and be at greater risk of an There are data which show that of both in the and the is a major with the use of these More is likely to be by people to at the time and in than be by on higher The and and by time of that at percent of occurs in the two either of from the Using the data it is possible to there are particular populations and that be the of both public and education programs. education programs are not the only of the public health approach to melanoma control. They in with change can be as a change to the or of a that may to in or other in a cancer control program. change can be of education programs, they In examples of change include or to make them more and thus more difficult for to of change in melanoma control are shown in In there are many that can be used to primary prevention education include primary and health and who provide and for and in to the and with on from are for for These include at and other people may large exposure to sunlight. can be by There are many other and which both and can be The and to which programs will be developed are by the of melanoma in a particular population and the of the and that that population at in the or be to excessive sunlight exposure during or to In countries to the susceptible populations year in the sunlight (e.g., and more programs be to to change the on a basis in a large number of the In the early a program a a a in Australia. This program of on and and other and the very positive approach that can be taken to a primary prevention program. In to control programs, the is to exposure the programs can people that being or on is and they can concentrate on people to the the risk of melanoma. can be in a very positive and that is more likely to have them noticed and by the various populations than a program that people not to do all the time and be seen as a or in with number an As the time between sufficient sunlight exposure to develop a melanoma and its clinical may be many decades. Thus, of primary prevention programs to include of that can be used to within the relatively (Table include behavioral variables that which the population at These include and the of sunlight and melanoma, as well as and of people in various of these behavioral variables can be used as rather than from behavioral research used to the and of these variables for of education programs can be used to the outcome. include a reduction in the of people with over an increase in the of people with of melanoma and its and an increase in the of individuals who that they are at of these are important in the process that in a change in The study of in mentioned above is an of an measurement of an effect of such programs. that can be used to early of programs include the of people using and the and particular the the around the of the and the using They can be in a of from to at of individuals they are that they are under The of melanoma with sunlight exposure sufficient to cause points to another relatively that can be that in populations at is a of the to which tissue the has been to of the of the population that during over or on can be It is to for and when between in populations at that it is to change in the a than to long-term change a The is more with of being to of is an important of the Thus, of change in must be over time to that they have and that they in the an increase in with decreasing from and The long-term of success of a reduction in mortality as a of a reduction in the incidence rate of melanoma. The value of and of melanoma incidence data be in the The data include not only invasive melanoma in melanoma as there is no in level melanoma the reduced incidence of invasive then this success of an early detection program rather than a primary prevention incidence data will to for a reduction in the incidence rate in the before the onset of a reduction in the incidence rate for the The public health approach to disease control is now being at with for many Melanoma an of a disease impact can potentially be influenced by this approach. the available both the cause of melanoma and its in at risk, the two approaches of primary prevention and early detection However, there are still in and is in the development, delivery, and assessment of of these programs. Melanoma control programs need to be by research data to the the population at and education and their effect in both the and or of data may cause and programs to or at Nevertheless, with a basis there is to that both early detection and primary prevention have the to be In countries melanoma incidence and mortality rates are public health there is for on the public health approach to melanoma control.