Looking Back to the Future – Acne

The purpose of this review is to try to stimulate more clinicians and scientists to become significantly involved in the investigation of acne, at all levels. Although along with eczema and psoriasis acne is the most commonly seen inflammatory skin disease, there are disproportionately fewer colleagues investigating acne than eczema and psoriasis. There are several likely explanations. At the research level, a biopsy of eczema and psoriasis immediately provides diseased tissue, whereas in acne a skin biopsy will require multiple serial sections before the pilosebaceous unit is fully identified. At the clinical level, the initial success of oral isotretinoin made clinicians, including our own group, believe incorrectly that there is little justification in the further investigation of acne. Whatever the precise reason or reasons, the numbers of active clinicians and scientists working on acne has declined in contrast to those clinicians and scientists working on other common inflammatory dermatoses.Nevertheless, much has been done to help acne patients, but much more research is required. I hope that this short paper will convince the readers of this need. Acne is a fascinating disease.Future researchers will fail to achieve their goals if they ignore the past. History tells us many lessons and techniques, both in the clinic and in the laboratory. It is so easy to forget the past, particularly with all the modern technologies available, and so ignore many of the great contributions. Yet with modern information technology it is now so much easier to surf the past.I wish to stress that without much of the solid foundations laid by predecessors and contemporaries I would not have been in a position to carry out our own research in Leeds. Teamwork within the department and beyond is essential. I have been extremely fortunate to have the opportunity of working in Leeds with an excellent group of clinicians, microbiologists, immunologists and biochemists, as well as worldwide with very many pleasant and intellectually stimulating colleagues. Acne folk are a good bunch of people.I have chosen to divide this paper into several sections. Firstly I will deal with a backward/future look at the aetiology of acne, which will cover the mechanisms involved in seborrhoea, comedone formation, Propionibacterium acnes colonisation and inflammation. I will also discuss clinical types, treatment of acne and clinical trials.The hormonal dependency of the sebaceous gland, especially the pilosebaceous gland, was established in particular by the groups around Montagna, Ebling, Strauss, Pochi and Shuster [1, 2, 3, 4, 5, 6]. The human endocrine studies of Strauss and Pochi confirmed the androgen dependence of the sebaceous gland [4, 5]. This was supported by the animal studies of Ebling [7]and his co-workers. Thody and Shuster [6]extended both the human and animal understanding of the hormonal control of the sebaceous gland.With time, it was apparent that the endocrine, especially androgen, control of the sebaceous gland was complex. It became obvious that the only way forward to better understand the detailed androgen control at the cellular level was to culture sebocytes. Akamatsu et al. [8], Zouboulis et al. [9], Guy et al. [10, 11]and Thiboutot et al. [12]have led the way on this front. In particular, Zouboulis has shown the close interlink between sebocytes and keratinocytes demonstrating an intricate local control of androgen metabolism within the skin.Many questions remain to be answered – some of which are: 1 Why are not all sebaceous follicles involved in acne? In the clinic, we see many patients with acne on the face but not on the trunk, or vice versa; sometimes the acne may be localised to one particular area of the body as in naevoid acne. 2 Can we in vivo identify with certainty acne-prone follicles? If so, is there a need to examine in vitro the growth control of such follicles in comparison with non-acne-prone follicles? 3 Are more specific androgen enzyme pathways likely to be found? The work of Thiboutot and Zouboulis has shown the sebaceous gland specificity of 5α-reductase type 1 [12]. Such developments have pharmacological implications in that specific inhibitors may be developed which could switch off the seborrhoea.The animal studies by Nikkari [13]demonstrated the unique composition of human sebum. The animal experiments of Kligman and Katz [14]demonstrated the likely comedogenic nature of certain lipids, in particular certain fatty acids, whereas deficiency of linoleic acid in sebum as shown by Wertz et al. [15]may help to explain in part comedogenesis. For a long time, lipid composition remained much neglected. Recently, Thielitz et al. [16]have sampled lipids from different follicles using skin surface biopsies and have shown differing lipid composition from different follicles and the direct effect of different topical therapy on lipid composition. This experiment highlights an important aspect of acne that is frequently forgotten – which is that acne is a disease of specific follicles.Many questions need to be answered, and these include: 1 Does individual follicular lipid (and cytokine) composition help to explain the propensity of a particular follicle to become a microcomedone? 2 Does individual follicular lipid composition help to explain ductal colonisation with P. acnes?Thus, in the future, we should compare lipid composition from individual sebaceous glands and follicles. This work would require sophisticated sampling techniques as well as scaled-down, very sensitive analytical techniques.The elegant experiments of Strauss and Kligman [17]were pivotal in showing that the microcomedone is one, if not the first, pathogenetic feature of acne. The further studies of Plewig et al. [18]and Holmes et al. [19]showed that even normal-looking skin from acne patients may already be developing very early microcomedones; this was further supported by the work of Cunliffe et al. [20]using in situ hybridisation techniques. This observation is of clinical significance since it stresses the need to apply topical therapies, not just to acne lesions but also to the normal-looking skin of the acne-prone areas.Aldana et al. [21]have also provided evidence that the pilosebaceous follicle may be involved in a cycling process in both healthy follicles and in comedone formation. It was over 50 years ago that the hair cycling was described, and yet only recently has it been proposed that pilosebaceous follicles in health and disease may also undergo some form of cycling. This possibility of follicular cycling may explain the resolution of individual comedones. If one assumes that comedones are one of the earliest acne lesions, then if there is no spontaneous resolution of such lesions the faces of affected individuals would be a mass of whiteheads within a matter of years. Clinically this is clearly not the case. There has to be some form of spontaneous resolution of such lesions.What controls ductal hypercornification is complex. The early animal studies by Kligman and Katz [14]suggested that certain sebaceous lipid components may be a factor, and subsequently Wertz et al. [15]proposed that a low level of sebaceous linoleic acid may also be involved. Thiboutot et al. [22]have shown that 5α-reductase type 1 activity may be involved in the formation of comedones and the exciting in vitro production of comedones by Guy and Kealey [23]in the laboratory has shown that some cytokines such as interleukin type 1α are comedogenic whereas other cytokines such as epidermal growth factor will prevent comedone formation. The detailed electron-microscopic work of Knutson [24]suggested that bacteria are not involved in the initiation of acne. Time-consuming single-biopsy work from the laboratory of Holland also showed that P. acnes may be involved in the later stage of comedogenesis but not in the initiation. These studies again stressed the need for researchers to, when possible, investigate single follicles [25].The clinical benefits of topical retinoids on comedogenesis were the subject of a pivotal study by Kligman, Leyden and Plewig. This has led to a major role of the use of this class of compounds in acne.Many questions remain unanswered, for example: 1 What is the precise interrelationship between the individual factors (androgens, lipid composition, cytokines and bacteria) which contribute to comedone formation? 2 Is failure of adequate ductal corneocyte separation an important factor in comedone formation? 3 Is follicular cycling a real event or not? This hypothesis needs to be further addressed.Historically in the early 1960s there was debate as to whether or not P. acnes was even involved at all in the pathogenetic process – particularly since the work of Marples [26]and Leyden et al. [27]showed that many patients with acne had similar levels of surface P. acnes as did control subjects. However, Leyden et al. [27]also showed that it was only with those antibiotics which significantly reduced skin surface P. acnes that there was associated clinical benefit. The demonstration by Eady et al. [28]that in some patients the presence of resistant P. acnes (especially those receiving oral erythromycin) was associated with clinical failure added further support to the concept that P. acnes is likely to be the most important micro-organism involved in the inflammatory aetiology of acne. International collaboration with colleagues such as Dreno et al. [29], Gollnick and Leyden has shown that the increasing prevalence of resistant P. acnes is a worldwide problem. The group of Eady has also demonstrated that the resistance is mutational and is thus likely to be a persistent problem [30]. However, the relationship between P. acnes resistance and clinical failure is not a straightforward relationship; some patients with microbiological resistance will still respond clinically, which emphasises even more so the multifactorial nature of acne.Given the fact that many control individuals have the same number of surface P. acnes as patients with acne raises fundamental issues in our understanding of the relationship between P. acnes and acne. What is likely to be more important is follicular colonisation. In 1978, Holland et al. [31]put forward what still remains a very acceptable hypothesis, which may explain the important role of the microbiological involvement of P. acnes in acne. They suggested that specific changes in the follicular micro-environment may allow follicular colonisation by this non-motile organism. This possibility was supported by much in vitro work and limited in vivo data. Biopsies of some follicles from acne patients are usually colonised but others are not.Which microbiological questions need to be addressed? 1 Why are some follicles colonised by P. acnes and others are not? 2 Which are the relevant pro-inflammatory mediators produced by P. acnes? 3 What is the precise clinical significance of P. acnes resistance? 4 What are the precise reasons for the development of P. acnes resistance? 5 How can P. acnes resistance be controlled and treated?Where possible, techniques should be developed and extended in order to allow sampling and analysis of specific follicles prone to acne compared to those which are not. This is true for sebaceous gland function, duct function as well as colonisation of P. acnes.Strauss and Kligman [17]demonstrated that the polymorphonucleocyte is an early cell in acne inflammation. This pivotal study was extended by Norris and Cunliffe [32], who developed a mapping technique for identifying very early acne lesions. The study demonstrated that in papules T helper cells are the first inflammatory cells and that in 80% of papules, comedones or microcomedones were a prerequisite for inflammation but duct rupture was not an early event. Further detailed studies by Layton et al. [33]using some mapping techniques showed that the inflammatory cell and cytokine profile in papules is that of a delayed cellular reaction to an antigen or antigens, the nature of which is yet uncertain.Issues have been raised as to whether or not acne patients may be immunological misfits. Shalita demonstrated that only in patients with grim disease was this the case.Thus, where shall we be looking in the future? 1 Timed biopsies are needed to understand better the sequence of events in pustules, macules, nodules and scars. 2 The relevant pro-inflammatory factors need to be ascertained. 3 How does therapy influence such events?Thus, any such future studies require: timed biopsies; detailed serial sections, and at a clinical level the assessment of the precise time course of acne lesions.This is an area where there has been surprisingly not much work – despite the fact that we can see the disease. In 1991, Pochi et al. [34]directed a delegation of colleagues to better define acne lesions. Most clinicians define acne as being mild, moderate or severe, inflammatory or non-inflammatory, but in clinical practice few clinics use adequate ways of grading acne. I really feel that this is a major mistake. Many of our colleagues in other specialities measure events such as pulse rate, blood pressure and the like, and yet as dermatologists, even though we can see what we are looking at, rarely do we use adequate ways of assessing skin disease. Some physicians do use pictorial grading scales, but still very few. I really feel this to be essential for the adequate management of acne patients in the clinic.Some physicians do not even use a lens to detect specific comedones which may influence the response to therapy. Although several groups, especially those of Kligman, Strauss, Plewig and Leyden, have demonstrated the important role of the comedone, a much better clinical description of the disease is required [34]. For example, Cunliffe et al. [20]showed that there are many sorts of comedones – sandpaper comedones, submarine comedones, macrocomedones, to name but a few – but rarely do clinicians identify such lesions which may have a major impact on disease management. It is also important to note that such lesions may have different pathogenetic features and patients prone to certain types of comedones need to be accurately identified because of potentially different mechanisms and different therapeutic needs.The psychosocial effects of the disease have been known for years. However, the recent work of Layton et al. [35]and Motley and Finlay [36]have identified the need and developed techniques to help the physician better assess this aspect of the patient’s acne.There is a great interest in the human genome. Blood samples are being investigated for potential disease-related genes. Unless the acne population from which the samples are taken is well defined, then such pivotal studies may be worthless.Acne patients should thus be classified according to age of onset, site of lesions, type of lesions, severity, proneness to scars and response to treatment.Lucky et al. [37]were responsible for an excellent study outlining the early onset of acne and demonstrating very clearly that the comedone was the earliest clinical feature, but there is virtually no work on the resolution of acne.The clinician also has a responsibility to help in identifying unusual cases which may give some useful aetiological clues. Fulton et al. [38]described gram-negative folliculitis. Several authors have described acne fulminans, but it was a larger study of Karvonen [39]which helped to redefine the problem better. More recently, Jansen et al. [40]provided a forum for debate on pyoderma faciale which they redefined as rosacea fulminans. In order for us to better understand the relatively uncommon disorders, perhaps there should be a data bank system, maybe organised by the American Academy of Dermatology and the European Academy of Dermatology, whereby information on such cases can be centralised.The clinician should also look out for atypical acne cases. Such patients could provide a wealth of scientific data; for example, patients with naevoid acne would allow the researcher to investigate why an individual has acne in one area and yet, given the same systemic background, hormonal and immunological, does not have acne at other sites. Again, such patients and their physicians may be willing for referral to a specialist centre.It is unfair to compare the robust nature of clinical trials performed now with those performed 20 years ago. Nevertheless, many early studies clearly showed the benefits of treatment. For example, Kligman et al. [41]were pivotal in laying down the important role of topical retinoids. Newer retinoid developments include new structures and reformulation of older molecules to enhance the benefit and decrease side-effects [42, 43]. Topical and oral antimicrobials have been reported by many colleagues such as Shalita, Leyden and Gollnick as well as Strauss and Stranieri [44]. Many others [45, 46, 47, 48, 49, 50], among them Strauss, Pochi, Leyden, Shalita, Plewig and Jones, identified the important benefit of oral isotretinoin in acne patients, initially those with severe acne as well as its benefit in atypical acne variants. Harms et al. [51]showed that there were important risk factors which related to relapse of acne after Roaccutane. The late 1990s was the era of the Cochrane guidelines and CONSORT [52], which should now be essential ingredients of clinical trials. These concepts are based on the need, quite rightly so, of double-blind randomised clinical trials. However, in no way should some of the studies performed 20 years ago and which do not match up to Cochrane guidelines and CONSORT, be invalidated. This is totally unjust. They were done and done well, as assessed by the yardstick of the time – but what about the future? There is much variation in clinical trial methodology, not just concerning acne but other dermatoses and other diseases as well. Vast amounts of money are spent, particularly by the pharmaceutical industry, in the design of clinical trials. There is a need to better standardise the assessment of the disease, side-effects of therapy, patients’ perception, statistical analysis and documentation. Thus, what is really required is harmonisation of acne clinical trials worldwide so as to reduce expense, avoid repetition of studies and allow better comparison between studies. The Executive Board of the EADV has given their support for this concept of harmonisation of acne trials, which requires cooperation between physicians, drug companies and the regulatory authorities.The number of papers on this subject can be counted virtually on one hand; thus, we know very little about resolution. What we do know? From the clinical studies of Goulden, in 7% of subjects the acne persists beyond the age of 25 years and there is often a family history, suggesting a genetic predisposition to persistent disease. Such patients are not hormonal misfits, and there is no excessive colonisation with resistant P. acnes compared to the adolescent The does not as the acne but what to comedone formation, ductal colonisation of P. acnes and the inflammatory response in acne resolution is totally can be in ways – not just resolution of the disease as a but also as resolution of individual lesions. The data cycling of the follicle in health and disease may help to explain the resolution of comedones. is an area out to be on the clinical of acne, much work is required. There are needs – ways of assessing of the disease – ways of assessing the psychosocial impact of the – better the of the – investigate the resolution of individual acne lesions and the disease – that when the laboratory a patient’s that the precise about the acne are well – clinical is is that much pivotal work has been but there is still much to be into clinical a good look at the may help to better on the more relevant questions for the wish to to all colleagues who have been involved in acne research and made to our understanding of the Cunliffe is the Cunliffe of the for his to clinical and