Correspondence 457 Table 2 Estimated NNTBs for narrowband ultraviolet B phototherapy – data on percentages reaching PASI 75 from two controlled studies (one with high PASI 75 response rate and the other with low PASI 75 response rate). In the absence of available placebo-controlled studies, ‘sensitivity analyses’, with low (3%, as in one RCT concerning etanercept), intermediate (10%) and high (19%, as in an adalimumab RCT) percentages of placebo-treated patients were conducted. For the purpose of these estimates, numbers assumed to have been treated with placebo were taken as the same as in the active treatment groups
Study reference
Regimen
Outcome timepoint
Percentage of treated patients who achieved PASI 75
Boztepe et al. 20067 Koek et al. 20098
3 · weekly therapy, starting dose 60–70% MED, 20% (reducing dependent on erythema) increments Variable regimens; 2 · to 3 · weekly, and MED-based and non-MED-based regimens
12 weeks 12 weeks 12 weeks Mean 12Æ8 weeks Mean 12Æ8 weeks Mean 12Æ8 weeks
92 92 92 41 41 41
Percentage of placebo patients assumed to achieve PASI 75
NNTB (95% CI)
3 10 19 3 10 19
1Æ1 1Æ2 1Æ4 2Æ6 3Æ2 4Æ4
(1Æ0–1Æ3) (1Æ1–1Æ5) (1Æ2–1Æ9) (2Æ1–3Æ2) (2Æ5–4Æ4) (3Æ1–7Æ5)
CI, confidence interval; MED, minimal erythemal dose; NNTB, number needed to treat to benefit; PASI, Psoriasis Area and Severity Index; RCT, randomized controlled trial.
bers needed to treat with one treatment to see an extra patient reaching a good response, such as PASI 75 (or PASI 90 or clearance), who would not have reached that response with an alternative therapy. Few studies have been done to allow such calculations. In the absence of any direct comparative study evidence we have to rely on indirect comparisons. Overall, there is no evidence that TNF-a inhibitor therapy ‘biologicals’ are any more effective than other treatments such as NB-UVB. The adverse effect risks of anti-TNF-a drugs are discussed by Dharamsi et al.1 and that these serious adverse effects are already becoming apparent, despite our relatively short experience of these therapies, is concerning. For now, we should continue to consider TNF-a inhibitor therapy for psoriasis only when other longer-established, safer and cheaper (important in ensuring we can afford to treat everyone appropriately) approaches have not been adequate. Hopefully, to generate really useful information that will guide clinical practice, future studies of TNF-a inhibitor therapy for psoriasis will either (i) only recruit patients who have failed to respond to other therapies (for whom comparison with placebo is reasonable) or (ii) compare these drugs with other effective treatments, not just with placebo. Department of Dermatology, Ninewells Hospital and Medical School, Dundee DD1 9SY, U.K. E-mail:
[email protected];
[email protected]
R.S. DAWE
References 1 Dharamsi JW, Bhosle M, Balkrishnan R et al. Using ‘number needed to treat’ to help conceptualize the magnitude of benefit and risk of tumour necrosis factor-alpha inhibitors for patients with severe psoriasis. Br J Dermatol 2009; 161:605–16. 2 Manriquez JJ, Villouta MF, Williams HC. Evidence-based dermatology: number needed to treat and its relation to other risk measures. J Am Acad Dermatol 2007; 56:664–71.
3 Chaudhari U, Romano P, Mulcahy LD et al. Efficacy and safety of infliximab monotherapy for plaque-type psoriasis: a randomised trial. Lancet 2001; 357:1842–7. 4 Saurat JH, Stingl G, Dubertret L et al. Efficacy and safety results from the randomized controlled comparative study of adalimumab vs. methotrexate vs. placebo in patients with psoriasis (CHAMPION). Br J Dermatol 2008; 158:558–66. 5 Edwards SJ, Clarke MJ, Wordsworth S et al. Indirect comparisons of treatments based on systematic reviews of randomised controlled trials. Int J Clin Pract 2009; 63:841–54. 6 Bansback N, Sizto S, Sun H et al. Efficacy of systemic treatments for moderate to severe plaque psoriasis: systematic review and metaanalysis. Dermatology 2009; 219:209–18. 7 Boztepe G, Karaduman A, Sahin S et al. The effect of maintenance narrow-band ultraviolet B therapy on the duration of remission for psoriasis: a prospective randomized clinical trial. Int J Dermatol 2006; 45:245–50. 8 Koek MB, Buskens E, van Weelden H et al. Home versus outpatient ultraviolet B phototherapy for mild to severe psoriasis: pragmatic multicentre randomised controlled non-inferiority trial (PLUTO study). BMJ 2009; 338:b1542. 9 Gottlieb AB, Matheson RT, Lowe N et al. A randomized trial of etanercept as monotherapy for psoriasis. Arch Dermatol 2003; 139:1627–32. Key words: antitumour necrosis factor-a therapies, biological, placebo, treatment comparisons, ultraviolet B Conflicts of interest: None declared.
Iatrogenic melanoma. Comment on: Melanoma epidemic: a midsummer night’s dream? DOI: 10.1111/j.1365-2133.2009.09556.x MADAM, We read with great interest the recent article about the controversy over the explanation of the melanoma epidemic.1 The past decades have witnessed a substantial increase
2009 The Authors Journal Compilation 2009 British Association of Dermatologists • British Journal of Dermatology 2010 162, pp452–468
458 Correspondence
in the reported incidence of cutaneous malignant melanoma (CMM) without a proportional rise in melanoma mortality in most European countries. The paper suggests that the large increase is likely to be due to diagnostic drift which classifies benign lesions as stage 1 melanoma.1,2 Histology is the gold standard for the diagnosis of CMM, but the assessment of small and thin melanocytic lesions, that constitute a growing proportion of lesions submitted for histology, is problematic, and interobserver agreement is moderate at best.3,4 Histological indicators of malignancy have largely been derived from larger lesions, and it is unknown if they are equally applicable for small lesions. As the consequences of overdiagnosis are generally limited to a small local re-excision and increased patient stress, whereas underdiagnosis results in an increased chance of recurrence and death, judgement tends to be skewed towards malignancy.2,4 In our clinic, members of melanoma families have been under surveillance since 1981. In many of these families, a mutation (p16-Leiden) in the high-penetrance melanoma susceptibility gene CDKN2A has been identified.5 During surveillance of 37 families with a p16-Leiden mutation, melanomas have been diagnosed in 105 genetically tested relatives, 12 of whom (11%) were noncarriers. These 12 noncarriers had a total of 13 melanomas. As part of a study on the effect of surveillance (manuscript in preparation) the slides of 126 melanomas were reviewed. These consisted of all in situ melanomas (n = 63), and invasive melanomas with missing data or of a nonsuperficial spreading histological type (n = 52) that had been diagnosed in mutation carriers within these 37 families. All melanomas of the 12 noncarriers that were available for histological review (seven in situ and four invasive melanomas from 10 patients) were added to the set. Slides were revised by a pathologist who is a member of the Dutch melanoma panel (W.J.M.). Revisions were performed blinded for the patients’ mutation status. After disclosure of the mutation status a disproportionately high proportion of (in situ) melanomas reclassified as benign melanocytic lesions turned out to be cases of noncarriers. Eight (seven in situ and one invasive) of the 11 melanomas of noncarriers were reclassified as benign (73%), compared with only 13 of the other 115 cases (11%). In seven of 10 mutation-negative relatives a history of melanoma was therefore not confirmed. These results touch on two important issues. Firstly, the value of genetic testing for CDKN2A mutations has been discredited because of a reported increased melanoma incidence among mutation-negative relatives.6 Our data show that overdiagnosis may account for a significant proportion of this observation. Secondly, increased screening and surveillance of individuals with a low a priori melanoma risk may result in removal of increased numbers of small and histologically equivocal lesions, some of which will be overdiagnosed as cancers and (especially in the case of individuals with a single relative with melanoma) will contribute to the chance of an inappropriate picture of familial clustering.
Departments of Dermatology and Human & Clinical Genetics, Leiden University Medical Center, Leiden, the Netherlands *Department of Pathology, Vrije University Medical Center, Amsterdam, the Netherlands Correspondence: Nicole Kukutsch. E-mail:
[email protected]
J.I.
RHEE W.J. MOOI* N.A. KUKUTSCH F.A. DE SNOO W. BERGMAN VAN DER
References 1 Levell NJ, Beattie CC, Shuster S, Greenberg DC. Melanoma epidemic: a midsummer night’s dream? Br J Dermatol 2009; 161:630–4. 2 Swerlick RA, Chen S. The melanoma epidemic. Is increased surveillance the solution or the problem? Arch Dermatol 1996; 132:881–4. 3 A nationwide survey of observer variation in the diagnosis of thin cutaneous malignant melanoma including the MIN terminology. CRC Melanoma Pathology Panel. J Clin Pathol 1997; 50:202–5. 4 Brochez L, Verhaeghe E, Grosshans E et al. Inter-observer variation in the histopathological diagnosis of clinically suspicious pigmented skin lesions. J Pathol 2002; 196:459–66. 5 Gruis NA, van der Velden PA, Sandkuijl LA et al. Homozygotes for CDKN2 (p16) germline mutation in Dutch familial melanoma kindreds. Nat Genet 1995; 10:351–3. 6 Kefford R, Bishop JN, Tucker M et al. Genetic testing for melanoma. Lancet Oncol 2002; 3:653–4. Key words: CDKN2A, incidence, melanoma, overdiagnosis, stage Conflicts of interest: none declared.
Dermoscopic rainbow pattern in non-Kaposi sarcoma lesions – reply DOI: 10.1111/j.1365-2133.2009.09554.x MADAM, We read with great interest the correspondence by Va´zquez-Lo´pez et al.1 regarding the dermoscopic observation of the rainbow pattern in non-Kaposi sarcoma (non-KS) lesions. We would like to emphasize that our previous study was based on seven patients with KS and 63 patients with various other cutaneous vascular tumours and nonvascular tumours.2 In our study, we did not observe the dermoscopic rainbow pattern in the non-KS tumours, indicating that the rainbow pattern is highly specific for KS lesions. However, we do not in any way claim that the dermoscopic rainbow pattern is completely specific for KS, and it is possible that non-KS lesions may produce the rainbow pattern under dermoscopy. In their correspondence, Va´zquez-Lo´pez et al. showed three possible images of the rainbow pattern in three different skin lesions: nail melanoma, stasis dermatitis and lichen planus. We find it surprising that the rainbow pattern could be seen in melanoma lesions. It is known that numerous studies have been performed on the dermoscopic features of melanoma, 2009 The Authors
Journal Compilation 2009 British Association of Dermatologists • British Journal of Dermatology 2010 162, pp452–468
