Treatment of multiple, multiform actinic keratoses on the head with imiquimod 5% cream

ARTICLE

Auteur(s) : Eggert Stockfleth¹, Claas Ulrich¹, Bernhard Lange-Asschenfeldt¹, Hans-Joachim Kremer², Ulrike Drecoll¹, Joachim Maus³, Joachim Röwert-Huber¹

¹Skin Cancer Centre Charité, Clinic of Dermatology, Venereology and Allergology, Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany
²Medical Writing Service, Alemannenstraße 101, 79117 Freiburg, Germany
³Meda Pharma GmbH & Co. KG, Benzstr. 1, 61352 Bad Homburg, Germany

accepté le 31 Janvier 2009

In the US, imiquimod has already had marketing authorisation for the treatment of actinic keratoses (AK) since 2005. This decision was primarily based on vehicle controlled trials which investigated twice or thrice per week application of 5% imiquimod cream over 16 weeks [1-3]. However, clinical research with imiquimod in AK went on, leading to a modified treatment scheme, namely the application of one or two courses of treatment (COTs) [4]. One COT consists of a 4 week treatment period with 3 applications per week, followed by a 4 week treatment pause. A second COT may follow if complete lesion clearance was not achieved. During the treatment pause, the clinical inflammation induced by imiquimod subsides, but AK lesions continue to clear. An additional positive effect is the uncovering and subsequent eradication of invisible (subclinical) AKs in the treatment area. This schedule minimises drug exposure, yielding improved safety, tolerability, and thus, patient compliance [4]. As in the US scheme, the scheme with 1-2 COTS was studied in vehicle controlled trials [5, 6]. Imiquimod also received marketing authorisation for the treatment of actinic keratoses (AK) in the EU in 2006, primarily based on these two trials. However, all controlled clinical trials with imiquimod had restrictions concerning the type of AK lesions, the size of the treatment area, and other aspects to achieve optimal and standardised conditions for demonstrating a significant difference between active and vehicle. As translated now in the official labelling [7], they appear artificial and only partly reflecting the usual dermatological situations in daily practice. Indeed, many patients might also suffer from hyperkeratotic or hypertrophic AK lesions. Moreover, the restriction of the treatment area to a contiguous one of only 25 cm², as studied in all controlled trials with imiquimod, might be unrealistic in daily practice. Therefore, our aim was to confirm the efficacy of the European imiquimod regimen (1 or 2 COTs) under conditions much closer to dermatological practice than those of controlled trials.

The main results of this trial are already published elsewhere [8]. However, that manuscript neither presented data on multiple nor multiform AK lesions nor on the influence of locations nor on the relationship between local skin reactions and clearance. Hence, the present paper firstly focuses on these important aspects.

Methods

If, at any time, the investigator believed any lesion located within the treatment or surrounding area to be suspicious of invasive malignancy, the investigator was to remove the lesion according to standard medical practice.

Safety documentation and statistical evaluations were exhaustively described previously [8].

Results

Patient disposition, demographics, compliance

80% of patients had undergone a previous treatment for AK with the most frequently reported treatments being cryosurgery (32%) and surgical excision (28%). The mean number of AK lesions was 9.5. 523 of 829 patients (63%) had one or more hyperkeratotic or hypertrophic lesions at baseline (mean 4.4, median 3, range 1-40 lesions). For further details on background and baseline data cf. [8].

90% of patients applied at least 80% of the expected doses of study cream. Dosing compliance was similar for COT 1 and COT 2. Only 33 patients (4.0%) were considered noncompliant i.e. they had applied less than 60% of the expected doses of study cream. 134 patients (16.2%) took rest periods during the study, 122 during COT 1 (14.7%) and 23 (of 464, 5.0%) during COT 2 (of which 11 during both COTs). Patients taking a rest period during COT 1 and COT 2 missed on average 3.3 and 3.5 of the 12 scheduled doses, respectively.
Table 1 Premature discontinuations

Efficacy

Compliance rates between 60% and 100% were marginally, if at all, predictive for the outcome, however, the subset of patients (33) with a compliance below 60% showed a considerably decreased overall clearance rate of only 39.4%.

Females, patients younger than 65, and patients with skin type III or IV showed a slightly better overall outcome (table 2). Some imbalances, in particular the difference between male and female, might be explained by the localisation of the lesion (table 3). The overall complete clearance rate in patients with lesions confined to the scalp (66%) was lower than in patients with lesions confined to the forehead (81%) or the face (81%). This hypothesis is further supported by the observation that patients with multiple AK areas including the scalp had a worse outcome than patients with multiple affected areas excluding the scalp.

The number of baseline lesions was not predictive for the outcome up to a count of 8 (table 2). Patients with 1 to 8 baseline lesions showed complete clearance rates of approximately 80%. This proportion was reduced in patients with 9 to 16 baseline lesions to 61% and further to 47% in patients with more lesions. Nonetheless, even in patients with baseline lesion counts above 8 (252 patients), most lesions could be cleared. Non-existence of hyperkeratotic or hypertrophic lesions at baseline was only marginally predictive of a better outcome, or vice versa, patients with such lesions at baseline showed nearly the same complete clearance rate as those without. There was also no relevant difference in the time course of the lesion counts of any lesion and hyperkeratotic/hypertrophic lesions (figure 1). In fact, hyperkeratotic/hypertrophic lesons showed the same clearance kinetics as the non-hyperkeratotic/hypertrophic lesions (table 4).
Table 2 Complete clearance rates and subgroups

Safety and tolerability

The LSRs most frequently reported as severe by the investigator were erythema (32%), scabbing/crusting (28%), and flaking/scaling/dryness (15%, table 5). LSRs were more intense during COT 1 than during COT 2 [8]. LSRs were the reason for discontinuation in only 4 patients (2 classified as such in table 1, 2 further classified primarily as adverse events). Among the patients who discontinued the study due to adverse events, 2 complained of both LSRs and systemic reactions considered related to imiquimod, 4 complained of systemic reactions (fever, headache, myalgia, fatigue, shivering) considered related to imiquimod, and 11 experienced adverse events considered not related to imiquimod.

Cochran-Armitage trend tests indicated statistically significant relationships between severity of LSRs (except for flaking/scaling/dryness) and complete clearance rates, most pronounced for the COT 1 (table 6).
Table 5 Local skin reactions

Discussion

The general factors, female, age below 65, and skin type III and IV were correlated with slightly higher clearance rates. While the latter two might be obvious, the lower response in the male patients might have been related to the number of lesions, which might have been considerably higher in males, who have a bald head much more often than females. The area affected by solar damage might typically be much larger in males, leading to an increased number of lesions.

More interesting are the findings on the lesion count. Although our data indicate that the number of baseline lesions is inversely correlated with the clearance rate, this correlation is rather weak. Even patients with a high number of lesions (field cancerisation) benefit from the treatment with imiquimod. It should be noted that a considerable number of patients had a baseline lesion count above 16 and many of them could be totally cleared. At least, the number of lesions could be dramatically reduced in these patients.

With respect to hyperkeratotic or hypertrophic lesions, our data indicate that patients with such lesions have somewhat lower clearance rates than those without (66 vs. 75%). However, our data suggest that this lower success rate is not caused by the type of lesion (figure 1). This graph shows that hyperkeratotic or hypertrophic lesions may be treated with imiquimod as well as normal AK lesions.

The overall complete clearance rate of 68.9% found in our study was slightly above the estimates recently reported from vehicle controlled studies using the same regimen [5, 6]. Such a pattern, namely a better outcome in an uncontrolled setting than in a placebo-controlled setting, is often seen in many therapeutic areas and might be to some extent explained by the higher scepticism of investigators and patients when faced with the possibility of having been treated with placebo. Other important aspects may be more stringent selection criteria used for vehicle controlled trials. However, many of the relinquished selection criteria of our study as compared to vehicle controlled studies were expected to work rather in the contrary direction: We set no upper limit for the number of AK lesions, while the above mentioned studies allowed only 8 [5] or 9 [6] baseline lesions maximally. Furthermore, both controlled studies excluded patients with hyperkeratotic or hypertrophic AK lesions, which we included. Our data indicate that these factors cannot explain the higher success rates in our study.

Local skin reactions are common during treatment with imiquimod and are an indication that there is an increase in the cutaneous immune response at the site of treatment, triggered by increases in activated dendritic cells and CD4+ and CD8+ T cells [9]. Earlier studies have associated an increase in LSR intensity with higher complete clearance rates of AK lesions [1, 2, 5, 6, 9]. Exactly this relationship was carefully explained to all investigators in investigators’ meetings before start of the trial. This might have contributed not only to the high clearance rates, but also to the low drop-out rates due to adverse events. In fact, the rate of discontinuations during this study due to local skin reactions (4 out of 828 = 0.5%) was very low when compared to vehicle-controlled studies with the same regimen (4 out of 252 imiquimod recipients = 1.6% [5, 6]) or compared with other regimens investigated in earlier studies (3.9% of the imiquimod recipients [10]). We considered essential for the therapeutic success that physicians carefully explain to their patients the positive relationship between LSRs and clearance rate before starting therapy with imiquimod. This is important for achieving high compliance and subsequently optimum therapeutic success.

Future studies in AK should allow for a stratification of the long-term risk by the type of AK lesion. Such classification of lesions should, however, be based not only on clinical features such as hyperkeratotic or hypertrophic lesions, but also on histological characteristics of the most suspicious lesions. An ideal histological classification system should incorporate qualitative as well as quantitative aspects. A respective proposal considering all AK lesions as early in situ SCC because of the presence of atypical keratinocytes in all AK lesions has recently been made [12].

Acknowledgement

The following investigators recruited patients: Eggert Stockfleth, Berlin, Thomas Arnold, Geithain, Thomas Baur, Berlin, Babette Boehr, Oranienburg, Linda Duhn, Bernau, Olga Dürpisch, Frankfurt/Oder, Birgit Göhre, Templin, Steffi-Kathrin Grelke, Lübbenau, Michaela Jumar, Magdeburg, Brunhild-Ch. Junge, Magdeburg, Ralf Struß, Berlin, Gunhild Kratzsch, Leipzig, Andrea Kreß, Altenburg, Beate Kuchheuser, Magdeburg, Meike Kuckert, Gräfenhainichen, Martin Kuppinger, Potsdam, Hans-Joachim Lüdcke, Potsdam, Christian Münzberger, Döbeln, Axel Mussmann, Wolfen, Kathrin Neubert, Burgstädt, Volker Neumann, Bad Saarow, Simone Neumann, Bernau, Reinhard Pettker, Berlin, Manuela Randow, Berlin, Katja Schubert, Dresden, Ulrike Schuchardt, Meißen, Margrit Simon, Berlin, Frauke Trautvetter, Dessau, Peter Uhl, Berlin, Gabriele Wolz, Hoyerswerda, Annemarie Zobel, Spremberg, Elke Auerswald, Hohenstein-Ernstthal, Rainer Burdinski, Paderborn, Manfred Derr, Mergentheim, Eberhard Dielmann, Limburg, Kay Dirting, Hanau, Karl-Armin Döhnel, Braunschweig, Jürgen Fried, Aschaffenburg, Matthias Gebhardt, Zwickau, Klaus Gissler, Darmstadt, Matthias Herbst, Darmstadt, Heinz-Jürgen Hübner, Paderborn, Claus Gruss, Passau, Sabine Kallenberger, Hannover, Frank Karches, Hannover, Marie-Francoise Kiesel-Mayerus, Höxter, Lieselotte Klose, Northeim, Anja Langrock, Wetzlar, Karin Lindner, Erfurt, Gisbert Paul, Gera, Henning Platschek, Hannover, Karin Prifert, Rudolstadt, Günter Reimer, Bad Homburg, Daniel Schaefer, Öhringen, Peter Samuel-Schleussner, Bad Vilbel, Heinrich Stirn, Paderborn, Gerald Wandel, Paderborn, Stefan Weidmann, Buchen, Wolfgang Weiß, Erfurt, Waldemar Zapf, Rudolstadt, Sven Zedlitz, Darmstadt, Ann Baumgartner, Freiburg, Gudrun Besing, Gauting, H.W. Beuschel, Schwabach, Ulf Detmar, Bad Wörishofen, Norbert Eich, Schweinfurt, Stephan Friedel, Freiburg, Alexander Glaessl, Ulm, Rainer Gollhausen, Dachau, Stefan Golsch, München, Manfred Groß, Bamberg, Martin Hahn, Rottweil, Torsten Hauschild, Rheinfelden, Thomas Hebel, München, Viktor Heimbuch, Wertingen, Mechtild Hofbauer-Schmelzer, Weil am Rhein, Claus Jung, Germering, Peter Kostka, Nürnberg, Christoph Kuhn, Günzburg, Irmela Martius, Forchheim, Volker Lungershausen, Singen, Margarita Meisel-Stosiek, Neumarkt, Lorenz Pfaff, Waldkirch, Georg Schuhmachers, München, Edgar Selvaag, Dachau, Ulrich Speer, Ravensburg, Thomas Titzmann, Augsburg, Stefan Uhlich, Kempten/Allgäu, Michael Weidmann, Augsburg, Cosima Weiß, Freiburg, Oliver Zimmer, Lahr, Magnus Bell, Andernach, Werner Bender, Wiesbaden, Werner Beyl, Koblenz, Irmgard Brändle, Stuttgart, Winfried Breustedt, Koblenz, Christoph Fedel, Böblingen, Vitali Franzen, Mayen, Klaus Fritz, Landau, Folke Habermann, Koblenz, Ulrich Heidbüchel, Mainz, Dieter Hirzel, Stuttgart, Klaus Jäger, Wörth, Stefan John, Püttlingen, Jürgen Kloos, Neuwied, Jürgen Knauber, Pirmasens, Martin Knüchel, Mannheim, Rainer Kopner, Köln, Hubert Krautheim, Kaiserslautern, Alfhild Nadji, Köln, Arndt Poswig, Köln, Kay-Peter Rehders, Wiesbaden, Peter Rein, Dillingen, Marion Sahlfeld-Frey, Backnang, Barbara Steimer, St. Ingbert, Christoph Trennheuser, Saarlouis, Klaus Trinkaus, Kaiserslautern, Roland Weber, Koblenz, Hanspeter Welters, Neckargemünd, Stefan Altmeyer, Oer-Erkenschwick, Theo Bergenthal, Iserlohn, Joachim Bockhorst, Dülmen, Jörg Bohmeyer, Lüdenscheid, Thomas Dirschka, Wuppertal, Rolf Dominicus, Dülmen, Peter Dorittke, Mönchengladbach, Marco Fuchs, Kamp-Lintfort, Gerold Gerhards, Krefeld, Virginia Große, Kleve, Roland Hartwig, Wuppertal, Matthias Hoffmann, Witten, Franz Hübinger, Wuppertal, Tristan Ionescu, Krefeld, Bernd Kardorff, Mönchengladbach,Volker Kingreen, Hagen, Ulrich Klein, Witten, Arnold Köllner, Duisburg, Eugen Lang, Essen, Frank Moschner-Kunert, Herten, Claus Nüchel, Mönchengladbach, Herbert Onstein, Borken, Peter Pierchalla, Recklinghausen, Beate Pilz, Arnsberg, Andreas Reipen, Olpe, Wolfgang Riepe, Münster, Hermann Rudolphi, Oberhausen, Peter Ruppert, Bocholt, Sibylle Schaller, Mülheim, Dorle Schlebes, Bocholt, Hans-Heinz Schrooten, Moers, Markward Ständer, Bad Bentheim, Magdalena Uszynska-Jast, Wetter, Zafer Aboutara, Oldenburg, Manfred Börries, Hamburg, Stephanie Denzer-Fürst, Kiel, Peter Drieschner, Lüneburg, Hans Engelke, Husum, Gunther Frings, Hamburg, Peter Giesenberg, Varel, Andreas Götze, Norden, Hans-H. Hagemeier, Löhne, Günther Heidbreder, Norderstedt, Jörg Hermann, Bremen, Hartmut Kietzmann, Kiel, Maria Kosinski, Bielefeld, Oliver Kreft, Papenburg, Andreas Krisch, Norderstedt, Bernd Lange-Cordes, Papenburg, Werner Löntz, Bad Schwartau, Waltraut Maronde, Winsen, Hartwig Mensing, Hamburg, Andreas Montag, Hamburg, Kirsten Prepeneit, Buchholz, Bruno Schmolke, Hamburg, Stefan Schröpfer, Glückstadt, Rainer Sempell, Itzehoe, Anke Thierfelder, Lübeck, Volker Streit, Buchholz,Markus Wehmeier, Verl, Klaus Welp, Buxtehude. Prof. Dr. med. Eggert Stockfleth acted as co-ordinating investigator. Medical managers: Elena Rizova, MD, and Jens Bichel, MD. Study manager: Marion Carey-Yard. Monitoring, data management, statistics: Ecron, Frankfurt.

References

2 Korman N, Moy R, Ling M, et al. Dosing with 5% imiquimod cream 3 times per week for the treatment of actinic keratosis. Arch Dermatol 2005; 141: 467-73.

3 Szeimies R, Gerritsen M, Gupta G, et al. Imiquimod 5% cream for the treatment of actinic keratosis: results from a phase III, randomized.double-blind, vehicle-controlled clinical trial with histology. J Am Acad Dermatol 2004; 51: 547-55.

4 Salasche SJ, Levine N, Morrison L. Cycle therapy of actinic keratoses of the face and scalp with 5% topical imiquimod cream: an open label trial. J Am Acad Dermatol 2002; 47: 571-7.

5 Jorizzo J, Dinehart S, Matheson R, et al. Vehicle-controlled, double-blind, randomized study of imiquimod 5% cream applied 3 days per week in one or two courses of treatment for actinic keratoses on the head. J Am Acad Dermatol 2007; 57: 265-8.

6 Alomar A, Bichel J, McRae S. Vehicle-controlled, randomized, double-blind study to assess safety and efficacy of imiquimod 5% cream applied once daily 3 days per week in one or two courses of treatment of actinic keratoses on the head. Br J Dermatol 2007; 157: 133-41.

7 European summary of product characteristics of Aldara^®. E.g. Germany: www.fachinfo.de, e.g. United Kingdom: www.emc.medicines.org.uk

8 Stockfelth E, Sterry W, Carey-Yard M, Bichel J. Multicentre, open-label study using imiquimod 5% cream in one or two 4-week courses of treatment for multiple actinic keratoses on the head. Brit J Dermatol 2007; 157 (suppl 2): 41-6.

9 Ooi T, Barnetson RS, Zhuang L, et al. Imiquimod-induced regression of actinic keratosis is associated with infiltration by T lymphocytes and dendritic cells: a randomized controlled trial. Br J Dermatol 2006; 154: 72-8.

10 Hadley G, Derry S, Moore RA. Imiquimod for actinic keratosis: systematic review and meta-analysis. J Invest Dermatol 2006; 126: 1251-5.

11 Röwert-Huber J, Patel MJ, Forschner T, et al. Actinic keratosis is an early in situ squamous cell carcinoma: a proposal for reclassification. Br J Dermato 2007; 156 (suppl 3): 8-12.