MALIGNANT MELANOMA ---------------------------------------- A recent male patient presented with a 1 cm pigmented lesion on the upper back. An excisional biopsy was done with 3 mm margins, and slide was read by Clay Cockerell as melanoma-in-situ with clear margins. Is it necessary to re-excise this lesion, and if so, what margins are recommended? James R. (Jay) McCarty, M.D. -------------------- Remember that the traditional method of examining parafin embeded excisions sent to pathology laboratories examines less than 1% of the true sugical margin. June Robinsons studies indicates that most lentigo malignas (malignant melanoma-in-situ) are excised completely with a 0.6-0.9 cm margin. Mohs was used to check the section adaquacy. Parafin sections were also used. Our data at Tulane using the Mohs terchnique with parafin sections confirms the above data. Most L.M. were excised with 0.5 cm margins but several large (2 cm wide) L.M.'s required around 1 cm. Since it is on the back where extra skin is not a problem, and to be conservative, I would rexcise with an additional 0.7 cm margin. Ask your pathologist to "cut-in" the entire specimen for histopath exam. Timothy Flynn, MD -------------------- I would re-excise with an additional 5mm margin. rick sharpe -------------------- For in-situ, I would get 5-10mm margins. Therefore I would re-excise this and get additional margins. Jerry Eisner -------------------- How does one know the entire tumor is only .25mm thick unless it is step sectioned? (I know, you'll say it already was) There are so many variables such as sex, location, duration,host immunity, etc. that one would have to invoke chaos theory to determine the outcome for this patient. Interferon and or melanoma vaccine therapy certainly could do little harm. If I had a melanoma of this size I think I would want more than just excision, but this is personal paranoia. Haines Ely -------------------- KC Smith asked about prognosis on a 3 cm x 2.5 cm x 0.25 mm superficial spreading melanoma. I think the prognosis would depend most heavily on the volume of the tumor that actually represents invasive tumor. If most of the surface area is in situ stuff he might not be as bad off as the total area calculation suggests. I presume that there is a central region that measures 0.25 mm and is therefore the invasive component? If the volume of this region is large, he probably does have a worse prognosis than would be indicated by thickness alone. Mark Naylor, M.D. -------------------- I have a 45 year old man with a 2.5 x 3 cm 0.25 mm thick superficial spreading melanoma. I got to wondering if the prognosis for about 7.5 square cm of 0.25 mm melanoma is more like that for a 1 square cm 1.875 mm thick melanoma than for the "less than 0.85 cm thick" melanoma quoted in the literature. Maybe this guy should be treated as an intermediate thickness melanoma and offered interferon? Derm Clin 1991 Oct;9(4):643-8 Friedman et al states that tumor volume less than 200 cubic mm had 91.4% 5 year survival, while volumes greater than this had 16.7% survival; and on multivariate analysis volume was more accurate than thickness as a prognostic indicator. KC Smith MD FRCPC -------------------- Several months ago I recall stumbling across an online melanoma prognosis program that allowed me to input data such as depth, ulceration, location etc. I can't seem to relocate it. Does anyone have a web address for this page/program or a similar one? Kip Cullimore, MD -------------------- The AAD sells it on a CDROM with back issues of the JAAD and some other goodies or as afreesatndingmodel. It uses the NYU prognosis model. I'm not aware of an online version, but would love to know if one exists. Daniel Mark Siegel MD -------------------- There is a worthwhile address at NIH for melanoma information for physicians. gopher://gopher.nih.gov:70/00/clin/cancernet/pdqinfo/soa/Melanoma_Physician Let me know if you have an difficulty reaching this address. Jim Rothschild -------------------- To the List: 1. What parameters do you use for referral of a patient with melanoma for evaluation for adjuvant treatment? 2. How do you follow patients with melanomas (in situ, Level I, Level II, Level III) -for the first year or two -for the next five years -for years 5-10 -after 10 years? Barbara R. Reed, MD -------------------- 1. What parameters do you use for referral of a patient with melanoma for evaluation for adjuvant treatment? * Anything over 0.75 mm thick or tumor volume over 100 cubic mm (which I estimate as area times max thickness) goes to the oncologist for a talk about Levamisole and alpha-interferon (the local oncologists are partial towards Levamisole). 2. How do you follow patients with melanomas (in situ, Level I, Level II, Level III) * everyone gets a TOTAL skin exam (scalp, between the toes, genitalia and everything else) as soon as primary Dx is made, looking for second melanomas. -for the first year or two q 6 months (q3 months if they wish) -for the next five years q 6 months (q3 months if they wish) -for years 5-10 q 6 months (q3 months if they wish) -after 10 years? q 6 months (q3 months if they wish) I tell them there is at least a 5% chance that they'll develop another primary melanoma sometime in their life (higher if there are dysplastic nevi, lots of moles and freckles, history of blistering sunburn ever (most have that Hx), or a FHx of melanoma. I tell them that it is THEIR responsibility to examine their skin EVERY month, and bring changes to medical attention RIGHT AWAY because the cure rate for these things is close to 99% if they are removed early, and the surgery is less onerous. I tell them they should be examined q 6 months FOR THE REST OF THEIR LIVES; and for the high risk people its is q 3 months and "call me RIGHT AWAY if you find something - don't wait for your next regular appointment. I give them the NCI melanoma brochure, and also a card with color photos illustrating the types of changes to look for (with photos of BCC and SCC added). I tell them that "There is no health food, vitamin, over-the-counter, 1-800 #, mail-order, naturopathic or homeopathic remedy which will help you, and money and time should not be wasted on those things. (I don't want someone to waste 3 months with that garbage while their melanoma grows from LMM to level 3!)". Kevin C. Smith MD FRCPC -------------------- This is what I do. Also for the group, a question along related lines: Is sentinel node biopsy considered "standard of care" for "intermedicate thickness" lesions in your area? What are the criteria used for proceeding with this? 1. What parameters do you use for referral of a patient with melanoma for evaluation for adjuvant treatment? Tumor thicker than 1mm gets sent to medical oncology for evaluation for protocols. Positive nodes go to the appropriate surgical for in-continuity dissection. 2. How do you follow patients with melanomas (in situ, Level I, Level II, Level III) Q 3 months for 1st year; Q 6 months for second year, annually thereafter. Daniel Mark Siegel MD -------------------- Dr. Smith's guidelines mirror mine, except.... The options (and the scientific data supporting them) is rapidly evolving for adjuvent Rx of melanoma. 1. The current literature suggests that HIGH DOSE alpha interferon (not low dose) is probably useful for certain patients at high risk of recurrence (i.e. very high risk primaries or after removal of a regional nodal met). 2. My read of the literature on Levamisole makes me reluctant to recommend this Rx at this time. 3. Melanoma adjuvent vaccines are showing promise. 4. A number of other immunotherapies/biologic response modifiers are on the horizon....I would keep a close eye on the literature. Rick Sharpe -------------------- BRESLOW THICKNESS Lesion thickness 5-year survival (local dz) <=3D 0.75mm 96% 0.76-1.49mm 87% 1.50-2.49mm 75% 2.50-3.99mm 66% >=3D4mm 47% Breslow A: Thickness, cross-sectioned area and depth of invasion in the = prognosis of cutaneous melanoma. Ann Surg 1970;172:902. THIN MELANOMA (LOCAL): EVALUATION, TREATMENT AND FOLLOWUP thickness margin melanoma in situ 0.5cm <1.0mm 1.0cm depth: to but not including fascia ELND not recommended lab evaluations only to confirm clinical suspicion follow up q6months for 2 years then yearly for life Early melanoma - NIH consensus conference. JAMA 1992;268;1314-1319. INTERMEDIATE/THICK MELANOMA (LOCAL): EVALUATION AND TREATMENT thickness margin lymph nodes 1.0-2.0mm 2cm SLNBx, ELND 2.01-4.0mm 2cm SLNBx=09 >4.0mm 2-3cm SLNBx =09 Balch CM, Urist MM, Karakousis CP, et al: Efficacy of 2-cm surgical = margins for intermediate-thickness melanomas (1 to 4mm): Results of a = multi-institutional randomized surgical trial. Ann Surg = 1993;218:262-267.=20 Balch CM, Soong SJ, Urist MM, et al: A prospective randomized trial of = 742 melanoma patients comparing the efficacy of elective (immediate) = lymph node dissection versus observation. Ann Surg 1996 in press. INTERMEDIATE/THICK MELANOMA (LOCAL): FOLLOWUP thickness followup 1-1.5mm CXR qyr 5 yrs h & p q6mo 3 yrs then yrly >1.5mm screen: chest CT scan & serum LDH (abd CT scan) CXR qyr 5 years h & p q6mo 5yrs then yrly labs to confirm clinical suspicion Rogers GS: Advances in diagnostic technique. J Dermatol Surg Oncol = 1989;15:605-607. John Hendrix, MD -------------------- On 9/2/96 Barbara Reed asked about adjuvant treatment of melanoma: I am assuming that you mean adjuvant chemotherapy in the sense that it is given during or after definitive therapy for the primary (surgical excision in this case) to people considered to be at high risk of recurrence (thick primary) in an attempt to prolong survival. Since there is no truly curative chemo (except potentially, some experimental regimens such as LAK treatment or experimental vaccines of various exotic types and a rare cure from standard drugs such as DTIC), the only other use of chemo is palliative. Since as far as I am aware there are no substantiated (by multiple similar reports) reports of effective use adjuvant chemotherapy for melanoma (to routinely and reproducibly improve survival), particularly in the setting of a Stage I primary (e.g., no known metastatic disease), although we obviously are in great need of such an option for people with thick primaries. Therefore, any chemo given as an adjuvant by an Oncologist would either be experimental or ad hoc (e.g., difficult to support in a strictly scientific sense other than as experimental). In answer to the question concerning how I follow melanoma patients, I will see them frequently if I think their chance of recurrence is substantial (usually those with a 1.5 mm primary or thicker) meaning every 3-6 months (and PRN of course) for the first 1-2 years, and usually every 6-12 months for up to 5 years, and then every year after that out to 10 years (after that, they are probably cures). Most of these visits will be a physical exam and history only. Lab tests: at *most* a yearly SMA-20, CBC and chest X-ray for the highest risk patients (>1.5 mm primary). The point here is to pick up a recurrence ASAP so that effective palliative measures can be instituted (e.g., to relieve bowel obstruction from a single met to prolong survival). Removal of single mets in other selected situations (single brain met, single lung met) might also improve life expectancy a bit, depending on the situation (might be worth it to the patient to have another 6 months). Early palliative removal of regional nodes can also prevent a diffult nursing situation from developing (e.g., ulceration, infection and pain). If patients have funny symptoms I would act as appropriate to the situation. I think that people who have a >0.76 mm primary should probably be seen at least once by an oncologist, or at least a primary care provider in addition to their dermatologist (this is all assuming they don't opt for experimental treatment protocols which will obviously change things). This will assist a rapid reaction to a rapidly life threatening cancer-related complication should it arise. We are better at picking up second primaries, which is one of the best arguments for dermatologists being prominently involved in melanoma follow up (as well as the best advice on prevention of second primaries). I am comfortable following thin primary patients by myself, which is advantageous in this era of cost-cutting, and might also be a good reason for other dermatologists to develop the necessary expertise to do the same. For Stage I patients with thin primaries (less than 0.76 mm), I see them in 6 months after definitive treatment is done and once a year for 10 years; no lab work up unless history and physical exam dictates the need for it. Obviously, if they are at high risk for some reason other than the single primary, such as in the case of familial melanoma, I would see them more often (probably every 6 months). I usually do make sure that they at least have a baseline chest X-ray to prevent an old granuloma from being mistaken for a met at a later date. And by the way, Diane, I only use sterile gloves for full-thickness procedures (with suture closures), otherwise "clean" technique with exam gloves. Mark Naylor, M.D. -------------------- I was also interested in the status of sentinel lymph node biopsy among the group. It is my understanding that the results with dye injection are very much dependent upon the experience of the surgeon. For that reason, we have not been doing sentinel LNBx much. However, it sounds like the use of a radioactive tagged material (I'm not sure what they are using) localized with a special G. counter is "very simple" and may be useful. Our surgeons are considering obtaining one of the counters (apparently not cheap) and using it alone or in conjunction with dye injection. Anyone in the group have experience / comments? Mark Crowe, MD -------------------- Adjuvent therapy (biologic response modifiers, NOT CHEMO) are showing promise for melanoma. In particular, high dose alpha interferon for patients at high risk of recurrence as well as melanoma vaccines for these same patients should be considered on an individual basis. As far as metastatic melanoma is concerned, I have several patients who were treated by Dr. Michael Atkins at New England Medical Center with IL-2 (no LAK or TIL cell Rx) who are in complete remission! One of these patients was advised by an Oncologist at another teaching Hospital to make the best of her last couple of months. That was more than 3 years ago...she remains completley disease free. Rick Sharpe -------------------- To the group- The attached is a chart from the Annals of Internal = Medicine this week - title: A Prognostic Model for Prdicting 10-Year = Survival in Patients with Primary Melanoma - Lynn Schucter et al, (U = of Pennsylvania and other centers) (AIM Vol 125 number 5 Sept'96 p369-375)=20 Only four parameters out of six examined were significant in predicting = survival. These were: Thickness (odds ratio 50.8): site of primary (odds ratio 4.4); age of = patient (ratio 3.0); and sex (ratio 2.0). They found this four-variable model "reduced the error rate of the = prediction of death by 50%" compared to using thickness alone. The attached file is in jpg form, one which allows significant = compression. Not everyone will be able to view it, and will have to = review the article itself. The file is less than 40 kb and should not = take too long to download, even for those who pay on-line charges. If = this is too cumbersome or seems inappropriate (eg - I did not ask the = author's permission to copy the chart) let me know. It seemed a good way = to disseminate interesting info by scanning instead of entering data by = hand. Jerry Eisner -------------------- To the group- The attached is a chart from the Annals of Internal = Medicine this week - title: A Prognostic Model for Prdicting 10-Year = Survival in Patients with Primary Melanoma - Lynn Schucter et al, (U = of Pennsylvania and other centers) (AIM Vol 125 number 5 Sept'96 p369-375)=20 Only four parameters out of six examined were significant in predicting = survival. These were: Thickness (odds ratio 50.8): site of primary (odds ratio 4.4); age of = patient (ratio 3.0); and sex (ratio 2.0). They found this four-variable model "reduced the error rate of the = prediction of death by 50%" compared to using thickness alone. The attached file is in jpg form, one which allows significant = compression. Not everyone will be able to view it, and will have to = review the article itself. The file is less than 40 kb and should not = take too long to download, even for those who pay on-line charges. If = this is too cumbersome or seems inappropriate (eg - I did not ask the = author's permission to copy the chart) let me know. It seemed a good way = to disseminate interesting info by scanning instead of entering data by = hand. Jerry Eisner I just do it at baseline, then only if indicated, never "routinely" eg. q year. KC Smith MD FRCPC -------------------- A more efficient way to handle this image distribution would be to post the Uniform Resource Locator, URL,(e.g. http://telemedicine.org/image.jpeg) in the text of the mail message. Some of the list members have low bandwidth connections to the Internet and are really bogged down by large attachments. Anyone who wishes to download the image can then simply double-click on the URL and get to the resource if their emailer supports object linking and embedding (OLE). If the mail package can't support OLE, they can simply copy the URL address and paste it into their Net browser. I am willing to post any graphics from this group on the IDS Web Server. Of course, the editors of the Electronic Textbook of Dermatology such as P. Marinkovich, A. Huntley, M. Naylor, et al. already have editorial access which allows them to post materials on the Internet Dermatology Society Web server through password protected file transfer protocol (ftp). They might be willing to post materials for you as well. Mark Naylor is particularly interested in melanoma and would be a nice contact person for posting this genre of table. As far as copyright goes, I spent last weekend speaking to this issue with an electronic guru from the NIH. He recommend that authors among us who have interesting information to share compile alternative data sets for electronic publication, other than those data which were used for other publications. Rhett Drugge, M.D. -------------------- It should be noted that the melanoma prognosis model reported in this week's Annals, and developed by the Penn Group (and Wally Clark), was developed to help the clinician by finding out which readily available variables could be used to predict survival. This group had published another model in the past, which was probably more accurate, but included several pathologic variables such as mitotic rate, infiltrating lymphocytes, etc., which are not often found in path reports. I guess I'm fortunate to have a local dermatopathologist (trained by Wally) who incorporates this previous model's variables in her reports of MM. I find it very useful. The new model is also useful, but it really amazes me that so many pathology reports apparently appear inadequate when MM is reported. The Annals article is worth reading thoroughly (unlike many articles in this esteemed journal!). Robert I. Rudolph, M.D., FACP -------------------- Apparently blood tests are not helpful at all, and CXR's should be considered in higher-risk cases only. Reference:Weiss, M., et. al.: Utility of follow-up tests for detecting recurrent disease in patients with malignant melanomas. JAMA 1995, Dec. 6; 274(21): 1703-5. Gene Sienkiewicz, M. D, -------------------- I have always eschewed CXR in following my thin or intermediate thickness melanoma patients, until acquiring a patient with the following history. He had a primary skin melanoma removed in Cincinnati about 35 years ago (thickness unknown) and four years later, on a routine "executive physical" CXR, a solitary lung metastasis was discovered. He underwent lobectomy and is in excellent health over 30 years later. Food for thought? Mark1105@aol.com -------------------- Food for thought indeed, and worth sharing, because I can add an almost identical anecdote --- except that it's been about 15 years with no recurrence after excision of a single lung met. Lab tests (LFTs, LDH, etc) have *never helped me, tho. Mike Eichwald -------------------- my husband has a similar case, but the CXR and the ensuing surgery was l2 years after the diagnosis, and now, over ten years later, the patient is still alive. Diane Thaler -------------------- Around 1980 the US melanoma incidence curve took a sharp turn upward. The corresponding melanoma death rate curve did not take such a turn. One can interpret this phenomenon in two ways: 1) The true incidence of a malignant (ie., potentially deadly) tumor increased at the exact same time that our ability to find and cure new cases increased dramatically; or 2) The criteria used by pathologists to diagnose melanoma changed at that time, permitting more non-potentially deadly lesions to be called melanoma. The second explanation seems a lot more plausible to me. Joe Clayman -------------------- WOW! I can barely believe some of the things I have heard from dermatologists on this list about melanoma. Go figure. Anyway, I guess its time to speak to some of these notions before somebody actually starts to believe them. Remember that the incidence of melanoma is the number of new cases usually expressed and the number per 100,000 per year. NOTION: Rise in melanoma incidence is due to diagnosis of non-melanoma. I think the American Society of Clinical Pathologists would take exception to this idea, although admittedly pathologists may be more likely to call questionable lesions melanoma than be sued later for failure to diagnose. This may be biasing the figure for earlier diagnosis, but many borderline lesions would probably go on to become fully malignant melanomas eventually anyway give more sun exposure (see earlier diagnosis below). On the other side of the coin, we have actually gotten more sophisticated in terms of recognizing melanoma look-alikes, e.g. the Spitz nevus, halo nevi, atypical nevi, etc. in recent decades. NOTION: Rise in melanoma incidence is due to earlier diagnosis. The fallacy of this is that it can only account only a temporary blip (for at most a few years), but absolutely can not explain the sustained rises we have seen over the last several decades All melanomas will eventually be diagnosed if they are melanomas, so earlier diagnosis, can logically only produce a temporary blip in the incidence figures, e.g. you are shifting some of the cases under the curve into earlier time periods but the curve would eventually level off or decrease slightly if this were all that was going on. NOTION: Rise in melanoma incidence is due to aging population (melanoma is a disease of old people). Much of the rise in incidence is actually occuring in younger individuals. The fact that melanoma may become the (according to NCI's SEER data) leading cancer-related cause of death in the 15-34 age group by the year 2000 is indicative of this though. I think anecdotally, all of us can attest to the fact that we are seeing plenty of melanomas in people under the age of 40. NOTION: Rise in melanoma incidence is due to increase in population. Just plain wrong. Incidence is per 100,000 people so whether the total population is increasing or decreasing doesn't matter, this is already figured into the number. NOTION: Rise in melanoma incidence is an artifact because the death rate rise has not been as steep. The death rate from melanoma is climbing, but the rise has not been as steep. In fact according to the SEER data summary for 1973-91, melanoma had the greatest increase in death rate for all cancers except prostate cancer. Unlike the incidence (which can't be explained on this basis--see above), this is almost entirely due to earlier diagnosis (which can explain this). Things like sentinel node biopsy and aggressive education of the population about the problem to bring down further the time between onset and diagnosis may blunt this rise further still (I sincerely hope it does). We as physicians should be proud of the fact that the increase death rate is not as steep as the climb in new cases. The plain truth folks, is that the climb in melanoma incidence is too massive to be easily explained away on the basis of faulty pathology or earlier case finding, or anything else other than the fact that there are many more melanomas per 100,000 out there than there used to be. We just can't stick our heads in the sand on this one people. We need to be out front, leading the way for better education, earlier diagnosis and prevention through smart sun exposure practices and widespread and aggressive use of high SPF sunscreens. In this way we can make a big, positive difference. Mark Naylor, M.D. -------------------- Incidence of the diagnosis of melanoma may depend on the subjective nature and specific criteria of histopathologic interpretation so I would rather use incidence of death from melanoma. What are the statistics for the current year and past years? Deaths per 100,000 persons in a specific age group per year is a very objective statistic which should be readily available from the CDC or the National Cancer Institute. If it is not, then I point to another example of bungling bureaucratic incompetence. Walter H. Wood, M.D. -------------------- Melanoma death rates have increased a consistent 2% annually (compounded) in the US since 1950 (when these were first collected) despite all we are doing to get the word to the public. I agree that this is the gold standard. Current rates are approx 2.3 / 100,000 (vs. approx 12.5 / 100,000 for incidence). Rates are higher in men than women. Studies have shown that the these rates are NOT related to changes in histopath dx criteria. DR DARRELL S RIGEL -------------------- On any criteria that one looks at: new cases per 100,000, total cases in the US, death rates, etc. there is a very clear increase in melanoma. Even with an overreading of pathology specimens because of medicolegal grounds, ther is no way one can interpret the data to show no increase in my opinion. I have been either reading my own slides or checking on my pathologist for over twenty years; there is clearly an increase in the number of cases on Maui in that time.Unfortunately, I have not been documenting the exact number, but I estimate an increase of 4 or 5 per year in 1975 to over 15 per year now. gary salenger -------------------- Several years ago, presented at an Am Acad of Dermatology meeting, NYU's data indicated that melanoma is not sun-related. With all the talk in RxDerm-L about sunscreens and melanoma-prevention, has new evidence appeared contradicting this concept? Philip Hughes, M.D. -------------------- As attached below, WH Wood asked a couple of days ago for the melanoma mortality rate for people aged 15 to 34. From Scotto et al. "Indications of future decreasing trends in skin melanoma mortality among whites in the United States", International Journal of Cancer, Vol. 49: 490-497 (1991): Table III (pg. 495) Average annual U.S. skin melanoma mortality rates and percentage change between 1975-79 and 1980-84 by sex and age group. White males under 30 1975-1979: 0.25 deaths per 100,000, 1980-84: 0.20 deaths per 100,000. white females under 30 1975-1979: 0.17 deaths per 100,000, 1980- 1984: 0.14 deaths per 100,000. Quoting from the paper, "Age-cohort analyses of skin-melanoma mortality rates (from 1950 through 1984) indicate that lifetime risks may have peaked for men born during the 1950s and women born during the 1930s, and provide early evidence that future trends in annual age-adjusted rates may bend downward, assuming that all other relevant factors associated with increased risk (e.g. sunlight exposure and surface levels of solar ultra-violet radiation) do not increase. Though age- adjusted rates increased by 2 to 3% per annum between 1950 and 1984, this was mainly due to upward trends among older men and women (over 40)." I would be gratified to think that this downward trend in younger age groups is due to dermatologists' increased vigilance and to our efforts to increase public awareness of the dangers of UV exposure. I certainly did not wish to create the impression from my previous post on this topic that I do not take seriously the public health hazard from melanoma. Marty Okun -------------------- I'm tired of reading the "lifetime risk" for melanoma. Give me deaths from melanoma per 100,000 per year for the 15-34 age group. Is it more or is it less? This should not be a subject of debate, it is a statistic. Walter H. Wood, M.D. -------------------- The histopathological criteria for melanoma have certainly changed, and, I dare say, they continue to. For example, a small, subtle melanoma-in-situ characterized only by a few nests of melanocytes would probably not have been called a melanoma several decades ago. We also know that the so-called active junctional nevus in most cases was what we call a melanoma-in-situ now (without hesitation). The converse is also true. We now know, for example, that "upward scatter" of melanocytes can be seen in nevi from genital skin. Do you believe that there were less melanomas in 2000 BC then in 2000 AD? Martin Reichel -------------------- There is overwhelming epidemiologic evidence that sunlight is the major etiologic factor in melanoma. For example, several studies in Europe and North America have shown an inverse relationship between increasing latitude and melanoma mortality [1-5]. In other words, as latitude decreases from the pole to the equator, the death rate from melanoma increases. The most likely explanation for this observation is that melanoma mortality (and incidence) varies with amount of UV energy reaching the earth's surface, which is known to be a function of latitude. The more UV, the more death from melanoma. In addition to this geographic evidence, melanoma is primarily a disease of lightly complected individuals, a fact which is probably due to the known UV-protective effects of melanin. In the Indian sub-continent, populated predominantly by individuals of skin type IV or V, the age-standardized incidence of melanoma has been estimated at around 0.2 per 100,000. In Australia by contrast, the incidence in the predominately skin type I-III populace living at roughly the same distance from the equator has been estimated at around 30 per 100,000, a difference in incidence of over two orders of magnitude [6]. In the U.S. (Los Angeles), rates in Americans of Japanese and Chinese ancestry (predominately type IV skin) were estimated at less than 1 per 100,000, while the rate in lighter complected Americans was 11-12 per 100,000 [7]. The anatomic predilection of melanoma for the upper back in men, and the leg in women also suggests that the sun exposed skin is what gives rise to these tumors [8]. As expected, the incidence on covered sites of the body is much lower [9]. I'm not familiar with anything coming out of NYU that seriously challenges this concept. Having said all that, Diane's right, sunlight is not the only etiologic factor for human melanomas. Another major known factor is genetic predisposition. One of the melanoma-predisposing genes that is responsible for familial melanoma has been characterized (the p16 kinase inhibitor or CDKN2A), but it is not the only one (there are still several that are uncharacterized). Another inherited problem that is known to predispose to melanoma (although they may not have the classic FAMMM syndrome) are defects in the retinoblastoma gene, which of course, also gives rise to retinoblastomas. If you think about it in theoretical terms, there are probably several (3-5 by one estimate) genetic defects in every malignant tumor, and it seems unlikely that melanoma would be an exception. That means that while the sunlight provides most of the genetic defects in melanomas, other things can too, e.g., an inherited genetic defect, or maybe environmental carcinogens or even potentially, a carcinogenic virus (such as a genital wart virus like HPV type 16 or 18). 1. Swerdlow AJ: Incidence of malignant melanoma of the skin in England and Wales and its relationship to sunshine. Br Med J 2(6201):1324-1327, 1979. 2. Magnus K: Incidence of malignant melanoma of the skin in Norway, 1955-1970. Variations in time and space and solar radiation. Cancer 32(5):1275-1286, 1973. 3. Eklund G, Malec E: Sunlight and incidence of cutaneous malignant melanoma. Effect of latitude and domicile in Sweden. Scandinavian Journal of Plastic & Reconstructive Surgery 12(3):231-241, 1978. 4. Teppo L, Pakkanen M, Hakulinsen T: Sunlight as a risk factor of malignant melanoma of the skin. Cancer 41(5):2018-2027, 1978. 5. Elwood JM, Lee JA, Walter SD, et al.: Relationship of melanoma and other skin cancer mortality to latitude and ultraviolet radiation in the United States and Canada. International Journal of Epidemiology 3(4):325-332, 1974. 6. Muir C, Waterhouse J, Mack T, et al., Cancer Incidence in Five Continents, in IARC Scientific Publications, No. 88, Anonymous, Editor. 1987, IARC: Lyon. 7. Whelan SL, Parkin DM, Masuyer E, Patterns of Cancer in Five Continents, in IARC Scientific Publications, No. 102, Anonymous, Editor. 1990, IARC: Lyon. 8. Crombie IK: Distribution of malignant melanoma on the body surface area. Br J Cancer 43(6):842-849, 1981. 9. Elwood JM, Gallagher RP: Site distribution of malignant melanoma. Candian Medical Association Journal 128:1400-1404, 1983. Mark Naylor, M.D. -------------------- Walter Wood asked about death rates from melanoma. NCI's SEER data is available over the WWW at http://www-seer.ims.nci.nih.gov/ if you go to the trouble of applying for a password and translating the database into something usable on your own computer (it is not very user friendly, and is currently targetted towards professional statisticians, but it is available). Death rates have gone from roughly about 1 per 100,000 population in 1950 to about 2.5 per 100,000 population in 1991 (Marty Weinstock's numbers actually). Only looking at death rates grossly underestimates the problem of course, and we are really concerned with how many people out there are getting melanomas, not just how many are dying from them. Mark Naylor, M.D. -------------------- There are many studies and facts to document and prove that the true incidence of Melanoma is rising. We (Al Kopf, bob friedman and I) reviewed these issues in this May's JAAD and I invite you to look at the article and hear you comments. Although several studies (Australia, Arizona, and ours in Manhattan) show that the rates are rising less rapidly ( and are almost flat in some studies) in YOUNGER age groups, the overall rate still continues to rise and cannot be explained by aging of the population. I apologize for the brevity of this response but I leave 7AM tomorrow for a 4 day trip to the islands (with lots of SPF 30 broad spectrum sunscreen, a brad brimmed hat and protective clothing).. Darrell Rigel -------------------- It is clear to me that the number of melanomas arising since "the ascent of man" from the cave dweller that humanity began as is constant. However, our ability to diagnose early, subtle melanoma-in-situ has improved dramatically. The melanoma epidemic is a result of an expanding population --an expanding population on earth, in dermatology, and in dermatopathology! Martin Reichel -------------------- The lifetime risks that I was throwing around are actually projections of the lifetime risk of a child born at that time (e.g., a child born in the year 2000 will probably have a 1 in 75 chance of getting a melanoma at some time in their lifetime). This is just a useful way of indicating how the incidence of melanoma is changing. The incidence of melanoma is very definitely increasing in my opinion, as is the death rate from melanoma (in all age groups). It is easier to project melanoma incidence than death rates because death rates may be influenced by earlier diagnosis and breakthroughs in therapy (I hope). Mark Naylor, M.D. -------------------- Thanks Marty for setting us straight about melanoma incidence in the young. My projections are based on the overall rate of all age groups, which is overestimating the risk in this age group if this trend is continuing. we can probably take some of the credit because of our efforts at sun protection and education (not only dermatologists, but many others of course). I entirely agree with you that this is evidence that an impact can be made and ought to spur us on to greater effort at prevention and education. Mark Naylor, M.D. -------------------- Maybe those nameless government "bureaucrats" were eavesdropping on Walter's complaint about inaccessible government statistics. Those of you who are interested and have been following the melanoma statistics thread (and have access to the www) might want to check out NCI's latest SEER data [1] on melanoma which can be downloaded and viewed with Adobe Acrobat (http://www-seer.ims.nci.nih.gov/Publications/CSR7393/index.html [download section 16 for melanoma statistics]). This digest does not require a password or any statistical analysis; it is a synopsis of interesting melanoma statistics viewed in several useful ways. I particularly recommend the last graph on page 14 which compares melanoma incidence by age and sex between the period 1973-75 and 1991-93 (the most recent SEER data). You will immediately see from this that the incidence of melanoma is increasing in all age groups, including white males aged 30-34 and under. For example, compare my estimates of incidence rates from this graphic (new cases of melanoma per year per 100,000 population) for white males and females aged 30-34: white males age 30-34 1973-75 1991-93 6.2 9.3 white females age 30-34 1973-75 1991-93 6.6 14.2 In other words, the SEER data indicates that melanoma is increasing in young age groups at present (all age groups in fact), contradicting the earlier Scotto conclusion that used data from the period between 1975-79 and 1980-84 [2] . Unfortunately, it looks like we have a lot of room for improvement. 1. Ries L, Kosary C, Hankey B, et al.,SEER Cancer Statistics Review, 1973-1993: Tables and Graphs. National Cancer Institute, Bethesda, MD, 1996. 2. Scotto J, Pitcher H, Lee JA: Indications of future decreasing trends in skin-melanoma mortality among whites in the United States. Int J Cancer 49(4):490-7, 1991. Mark Naylor, M.D. --------------------