- Email: [email protected]
Cycle therapy of actinic keratoses of the face and scalp with 5% topical imiquimod cream: An open-label trial
Cycle therapy of actinic keratoses of the face and scalp with 5% topical imiquimod cream: An open-label trial Stuart J. Salasche, MD, Norman Levine, MD, and Lynne Morrison, RN, BSN Tucson, Arizona Background: Preliminary studies indicate that topically applied immune response modifiers may be an effective and safe method of treating actinic keratoses (AKs). Objective: Our aim was to study the potential efficacy of topical 5% imiquimod cream in the treatment of facial or scalp AKs and improve the safety profile by using a novel “cycle” dosing regimen. Methods: This pilot study is an open-label trial that included 25 patients who had between 5 and 20 discrete AKs within a cosmetic unit of the forehead, scalp, or cheek. Treatment consisted of once-daily application of 5% imiquimod cream, 3 times a week for 4 weeks. to the entire cosmetic unit, followed by a rest period of 4 weeks. The cycle was repeated if any AKs remained after a complete 8-week cycle. A maximum of 3 cycles was permitted (24 weeks). Thirty-three sites in 25 patients were evaluated. Results: Compliance was excellent with a very tolerable safety profile. Complete clearing of all AKs was noted in 82% (27/33) of anatomic sites in 25 study subjects. Almost half the sites (15/33) were clear at the end of the first cycle. A “therapeutic interval” was noted during the rest period wherein clinical inflammation subsided but AKs continued to clear. An added effect was the uncovering and clinical appearance and subsequent eradication of incipient (subclinical) AKs in the treatment area. Conclusion: There was excellent compliance with the cycle therapy regimen. The observations and hypotheses made in this pilot study will be tested in controlled, randomized trials with larger study populations. The identification of a therapeutic interval may prove to be beneficial in formulating individualized dosing regimens. (J Am Acad Dermatol 2002;47:571-7.)
T
here has been a resurgent interest in nonsurgical, topical therapy for actinic keratoses (AKs). Recent publications emphasizing the status of actinic keratosis (AK) as an early stage in the biologic continuum leading to squamous cell carcinoma have kindled this attention.1-4 5-Fluorouracil has been the mainstay therapeutic agent in this area for decades.5-7 However, despite excellent efficacy with this drug, there is a high incidence of adverse effects, which has led to poor patient compliance and reluctance on the part of dermatologists to prescribe it. New formulations of 5-fluorouracil have been developed with the aim of lessening From the University of Arizona Health Sciences Center, Tucson. Funding sources: A grant-in-aid from 3M Pharmaceuticals. Disclosure: Dr Salasche is a medical consultant to 3M Pharmaceuticals. Reprint requests: Stuart J. Salasche, MD, University of Arizona Health Sciences Center, 1501 N Campbell Ave, Tucson, AZ 85724. Copyright © 2002 by the American Academy of Dermatology, Inc. 0190-9622/2002/$35.00 ⫹ 0 16/1/126257 doi:10.1067/mjd.2002.126257
these local skin reactions, and other products have been introduced to eradicate AK.8-12 Early studies with the topical immune response modifier 5% imiquimod have shown promising safety and efficacy profiles for treatment of AK.13-19 The rationale behind the use this new immunomodulatory topical therapeutic agent is that it is believed to stimulate and enhance the local host immune responses against viral infections and skin tumors.20-22 In this open-label study, a novel cycled approach was used in an attempt to maintain high efficacy while lowering the incidence and severity of skin reactions.
METHODS Study design This study was an open-label trial designed to examine the safety and efficacy of topically applied 5% imiquimod cream in the treatment of AKs of the face and scalp by using a cycle approach to therapy. Active medication was applied to the entire treatment area once daily, 3 times a week for 4 weeks. This was followed by a rest period of 4 weeks. The treatment area consisted of an entire cos571
572 Salasche, Levine, and Morrison
metic unit: the scalp, the forehead and temples, or the cheeks. For study qualification, a cosmetic unit was required to contain 5 to 20 discrete, nonhypertrophic AKs. Study participants were allowed to treat two cosmetic units if each satisfied this condition. For qualification in a specific cosmetic area, there could be no overlap in cosmetic units: for example, 3 AKs on the forehead and 2 AKs on the contiguous scalp. The scalp was defined as the area above the superior forehead crease, and the forehead-temple area extended below this crease to the eyebrow. Laterally, the forehead-temple area extended to the hairline above a line drawn from the lateral canthus to the superior attachment of the ear. The cosmetic junction lines of the lower eyelid, nose, lips, jaw line, and ear bound the cheeks. Study subjects were excluded if they had, within the prior month, used any topical medication within the treatment area that may have efficacy against AK such as topical retinoids, hydroxy-acids, 5-fluorouracil, or diclofenac or if they had received photodynamic therapy. Subjects were similarly excluded if they had had a dermabrasion, chemical peel, or laser resurfacing in the treatment area within 6 months. Immunosuppressed patients were excluded. The primary efficacy end point was complete clearance of AKs within the treatment area. No attempt was made to discriminate between AKs noted at baseline examination and AKs that arose in the intervening skin (subclinical AK) during treatment. If any AKs were present at the end a 4-week rest period, the entire area was re-treated for another cycle of 4 weeks, followed by another 4-week rest period. Each cycle then consisted of 8 weeks. A maximum of 3 cycles was permitted. If AKs were present after the third cycle (24 weeks), this was considered a treatment failure. The secondary efficacy end point was a 75% reduction in AKs over baseline. This secondary end point would correspond to a clinically relevant outcome. Histologic examination was not performed. As in the clinical situation, AK counts before, during, and after treatment were done by visual inspection and palpation. A lesion was considered an AK if the investigator would have ordinarily treated any individual lesion by destruction with liquid nitrogen. If the subject experienced moderate to severe local skin reactions during the active treatment part of the cycle, an early rest period would be initiated. The decision to impose a rest period was also influenced by the presence of symptoms that included pain, itching, burning, or tenderness. A rest period would also be imposed if systemic adverse events occurred. At the end of this rest period, a decision about whether to resume treatment would be made. This depended on
J AM ACAD DERMATOL OCTOBER 2002
the nature of the inciting event and/or whether AKs remained at the end of the rest period. The intent was to treat until there was complete clearing or a treatment failure at the end of 3 cycles. Participants were asked to apply the cream evenly over the entire cosmetic unit. Because one sachet covers approximately 20 to 30 cm2, patients were asked to use one sachet for the cheek, the forehead, and the anterior bald scalp. If the entire scalp were to be treated, two sachets would be required. Safety Study subjects were seen at baseline and then every 2 weeks until the study was complete, that is, either complete clearance of AKs or the end of 3 cycles (24 weeks). At each visit the following were done: a complete AK count, digital photography of the treatment area, evaluation of adverse local or systemic events, and an examination of the treatment site to assess local skin reactions. Local skin reactions were graded as mild, moderate, or severe. The signs that were assessed included erythema, edema, induration, erosion, ulceration, scabbing/ crusting, exudation, vesicles, and flaking/dryness/ scale. Symptoms of pain, burning, or itching were also quantified by using the same scale. Participants were questioned at each visit about any change in their health. They were specifically asked whether they had experienced any flu-like symptoms to include fever, chills, muscle fatigue, or a general feeling of weakness. Study population Twenty-five study subjects were enrolled and gave informed consent to the institutional review board–approved trial protocol. Thirty-three cosmetic unit sites on these 25 patients were included in the trial. Eight subjects qualified for 2 sites and chose to treat both areas. There were 23 men and 2 women. Anatomic site distribution was as follows: scalp, 13; forehead, 13; and cheek, 7. The average age of the study population was 73 years. This original group of 25 subjects and 33 sites constituted the basis of our intent-to-treat analysis, even though several subjects left the study for nonmedical reasons or did not comply with a sufficient number of doses to have qualified. Two subjects moved away from Tucson in mid-trial for personal reasons and were considered to have experienced treatment failures for intent-to-treat analysis purposes. Another subject withdrew consent after only 3 treatments because he was blind and his wife did not want to apply the cream to his scalp. This case was also considered a treatment failure.
J AM ACAD DERMATOL VOLUME 47, NUMBER 4
Salasche, Levine, and Morrison 573
Fig 1. A, Baseline treatment of forehead 3 times a week; actinic keratosis (AK) count ⫽ 19. B, At 2 weeks; AK count ⫽ 35. C, At 4 weeks; AK count ⫽ 40. D, At 6 weeks; AK count ⫽ 1. E, At 8 weeks; AK count ⫽ 0.
RESULTS Efficacy The primary efficacy outcome was 100% clearing of the AKs within the treatment area. On the basis of this criterion, there was a high clearance rate. For the intent-to-treat analysis, 82% (27/33) of the sites were
completely cleared (Fig 1). The 6 treatment failures included sites in two subjects who moved away in mid-trial and the one who withdrew consent after only 3 medication applications. Two other study subjects had accelerated brisk inflammatory reactions that necessitated an early rest period. In each
574 Salasche, Levine, and Morrison
J AM ACAD DERMATOL OCTOBER 2002
Table I. Total clearance by cycle: intent to treat sites (n ⫽ 33) No. cleared
%
15 12 0 27
46 36 0 82
Cycle 1 Cycle 2 Cycle 3 Totals
Table III. Total and average actinic keratosis counts from all 33 sites during first 8-week cycle Baseline Week 2 Week 4 Week 6 Week 8
No. of AKs
Mean
368 613 515 228 144
12 19 16 7 5
AKs, Actinic keratosis.
Table II. Time to complete clearance Cycle
Clear during treatment period Clear at 2nd wk of rest Clear at 4th wk of rest Total
First
Second
Third
2 2 11 15
2 2 8 12
0 0 0 0
of these cases, two sites were being treated. In both instances, at the end of the rest period, one anatomic site was clear and the other was not. Both subjects declined to resume treatment of the involved site, and each site was considered a treatment failure. A single subject went through three 8-week cycles and had persistent AK lesions. Among the cases of total clearance failures, one subject, the only one who completed the entire 3-cycle course, approached the 75% clearance required for the secondary efficacy end point. However, only 73% clearance was achieved because 3 AKs remained from a baseline count of 11. All subjects who experienced complete clearance did so within the first 2 cycles (Table I). Most experienced clearance during the rest periods (Table II). Table III summarizes the serial AK count data. Because complete AK counts of the treatment area were recorded at 2-week intervals, a time course of targetlesion response could be tabulated. The baseline AK average score of 12 AKs was reduced to an average of 5 AKs by the end of the first full cycle (week 8). The observation that AK counts increased in the weeks after initiation of treatment indicates that incipient, subclinical AKs were being uncovered and treated. The continued reduction in number of AKs during the rest period defines the concept of the “therapeutic interval.” As can be seen from Table IV, several study subjects who still had AKs at the end of 2 weeks into the rest period experienced complete clearance by the end of another 2 weeks without medication. Safety Overall, the dosing regimen was well tolerated by the majority of patients. Most of the local skin reactions were of the mild to moderate category. These
Table IV. Value of 4-week rest period Subjects with clearance during last 2 wk of rest who were positive after first 2 wk First cycle
9
Second cycle
8
tended to begin in the second to third week of active treatment and persisted through the fourth week when the rest period began. Resolution began promptly after cessation of active treatment. Severe reactions occurred in 5 patients, 4 of whom required early rest periods (Table V, Fig 2). All severe reactions occurred during the first cycle in the setting of an early, intense inflammatory reaction. In each case an accelerated response was noted within 1 to 2 weeks of initiating therapy. Erythema of the lesion and surrounding skin predominated and was accompanied by intense edema, induration, and enlargement of the AKs. This was followed by erosion and crusting of the AKs, leading to ulceration if the medication was not discontinued. Sensations of burning, pain, and tenderness accompanied these severe intense inflammatory reactions.
DISCUSSION A common approach to the treatment of patients with AK is to destroy small numbers of scattered, discrete AKs with liquid nitrogen.8,23 A different situation exists when there are numerous AKs in any given area or multiple areas of the exposed head and neck region and extremities. These may be best managed with a topical medication that eradicates not only the visible lesions but also the subclinical nascent lesions. Imiquimod is an imidazoquinolinamine, one of the new immune response modifiers. It has been approved by the Food and Drug Administration for the treatment of anogenital warts but has been studied as off-label treatment for several other skin indications including AK, basal cell carcinoma, and Bo-
Salasche, Levine, and Morrison 575
J AM ACAD DERMATOL VOLUME 47, NUMBER 4
Fig 2. A, Baseline; actinic keratosis (AK) count ⫽ 15. B, At 3 weeks: edema, erythema, crusting, tenderness; treatment discontinued; AK count ⫽ 30. C, At 4 weeks: resolution of adverse signs and symptoms; residual erythema. D, At 9 weeks: complete resolution of AKs and inflammation.
Table V. Summary of adverse events leading to early rest periods Patient No./Sex
11/F 21/M 23/F 24/M
Area
Forehead; cheek Forehead Forehead; cheek Forehead
Rest begun
3 wk 2 wk 2 wk 2 wk
Symptoms
Skin burning, severe erythema and erosions, fever, chills Mod erosions, erythema, burning Systemic malaise, aches; severe erythema, exudative crusts Burning, severe erythema
Skin type
II/Severe II/Severe I/Severe II/Severe
Mod, Moderate.
wen’s disease.13-19 The exact mechanism of action is not known but is likely related to the findings of preclinical and clinical research showing stimulation of both innate and acquired immunity. Monocytes, macrophages, and dendritic cells are induced to secrete cytokines that initiate a favorable inflammatory cascade, resulting in tumor necrosis. Cytokines interferon alfa, interleukin 12, and tumor necrosis factor ␣ are all generated locally. These in turn
activate natural killer cells of the innate immune system and induce the production of interferon ␥ that upregulates cell-mediated immunity.20-22 Recent data indicate that immune response modifiers act through TLR7 receptor (Toll-like receptor 7).24 Observations from this investigation with topical imiquimod indicate that when it is applied 3 times a week, all the AKs in the treatment area respond in a synchronized manner. Individual lesions become er-
576 Salasche, Levine, and Morrison
ythematous, edematous, and indurated. Subsequently, they become centrally eroded and crusted, and over the ensuing weeks, become smaller and ultimately disappear. In an earlier study, the peak inflammatory reaction took place between 4 and 6 weeks of treatment when a 3-times-a-week continuous, uninterrupted dosing regimen was used.15 The purpose of this open-label trial was to test the hypothesis that topically applied 5% imiquimod cream could be administered in a cycled fashion to attain the following results: (1) initiation of a therapeutic inflammatory response, (2) interruption of dosing at a time before the inflammatory reaction reaches a magnitude that would interfere or restrict daily activities, and (3) continued efficacy during the rest period as inflammation subsides. This preliminary open-label trial indicates that such a cycled approach of alternating 4 weeks of active treatment with rest periods of equal length may achieve high efficacy while minimizing adverse effects. This leads to high rates of patient acceptance, compliance, and satisfaction. This regimen is not to be confused with the “pulsed” therapy for AK with topical 5-fluorouracil in which topical application is done twice a day, one day a week until lesions are gone.6,7,25 Therapeutic interval: The primary benefit of cycled therapy Because we adhered to a rather rigid 4-week active intervention phase followed by a 4-week imposed rest period, we identified a “therapeutic interval.” During this gap (the rest period) between cessation of topical application of 5% imiquimod and restarting the treatment if any AKs remained, inflammation steadily diminished while the therapeutic effect continued. Treatment beyond the 4-week period probably would have, in many instances, resulted in unacceptable inflammation, unsightly appearance, and unscheduled cessation of treatment. Although inflammation had subsided sufficiently by 2 weeks into the rest period in most cases to have allowed the patient to restart topical application, there was continued improvement and clearing of AKs during the subsequent 2 weeks of the rest period (Tables II and IV). In the first cycle, 9 subjects with AKs who were 2 weeks into the rest period experienced complete clearance at the end of the 4-week rest period. In the second cycle, the same occurred with 8 additional subjects. Until sequential “mechanism of action” studies are done, we can only speculate that one of two possible events may explain these observations. It is possible that AK destruction is due to some critical molecular event that occurs early in the treatment cycle that programs the AKs for eradication. What
J AM ACAD DERMATOL OCTOBER 2002
we noted clinically is only the visible evidence of lesion regression. Alternatively, there may be an ongoing or self-sustaining therapeutic effect from the cytokine cascade initiated by the imiquimod. Although a 4-week rest period works well, it is still not clear whether there is an optimal active treatment time. For most patients in this trial, the 4-week, 3-times-a-week regimen was effective. A subgroup reacted very quickly and intensely, usually by 2 weeks. Although most of these rapid responders ultimately experienced clearance, they probably would have been much more comfortable with a slower, less intense response. We have identified potential risk factors for a local reaction, which include Fitzpatrick type I or II (severe to very severe) photodamage in the treatment area, many AKs within the treatment area, and female sex. Perhaps those with these risk factors should have a regimen of less frequent dosing (1-2 times per week), a shorter duration of active treatment (1-3 weeks), or both. This will be tested in future prospective, randomized trials. Furthermore, it may prove possible to titrate dosing during active treatment. When a predetermined level of inflammation or erosion/ crusting is attained, active treatment could be halted and the therapeutic interval could be initiated. In the clinical setting, complete clearance would not necessarily be the relevant end point. If only a few AKs remained at the end of a rest period, one might choose not to re-treat the entire area with subsequent cycles. Rather, focal lesions could be destroyed with cryotherapy or the patient could “spot” treat the remaining lesions with imiquimod cream. Subclinical AK By the time AKs develop, the skin in which they appear has been subjected to years of excessive sun exposure and photodamage, which results in varying degrees of hypopigmentation and hyperpigmentation, dermal thickening caused by solar elastosis, epidermal atrophy, and telangiectasia. In addition to the fully formed and clinically detectable AKs, the skin also manifests many undetectable clonally expanded foci of dyskeratotic keratinocytes. Although many of these “AKs-to-be” are probably destroyed by ongoing immune surveillance and local immune reactions, some eventually manifest as clinical AKs. One of the important salutary benefits of topical imiquimod treatment over cryotherapy of focal lesions is that not only are clinically evident AKs eradicated, but the surrounding photodamaged skin is also treated at the same time. In this study, the AK counts were noted to be elevated over baseline during the first cycle in almost every subject (Table III). Because the baseline AKs were merely counted
J AM ACAD DERMATOL VOLUME 47, NUMBER 4
and not sequentially tracked with a template during the study, it was impossible to discern the baseline lesions from the newly “unmasked” AKs. By the time the subclinical AKs were noted (by inference from an increased count over baseline), all the AKs within the treatment area were synchronized to the same stage of inflammation or resolution and had a similar clinical appearance. Finally, practicing clinicians should be cautioned not to construe the high complete clearance rates attained in this trial as the last word on this subject. It must be stressed again that this is an open-label pilot study whose results need to be corroborated by larger, prospective, randomized controlled studies. Such studies have been initiated. In a similar vein, it has not be ascertained whether the areas cleared by topical imiquimod in this clinical trial remain free of disease or whether new AKs will recur in the treated areas, and if so, when they will appear. To help answer these questions, the study group from this trial will participate in a follow-up study. The plan is to examine each patient every 3 months for 2 years. The time to appearance of any new AKs and the number of new AKs will be determined and subsequently reported. Conclusions The findings of this study must be evaluated cautiously. It is an open-label trial in a small number of study subjects. Safety, efficacy, and duration of efficacy must be corroborated by larger, randomized studies. However, cycle therapy and identification of a therapeutic interval may be a first step in allowing dermatologists to tailor dosing regimens to the specific patient characteristics according to the degree of sun damage and skin type. Furthermore, patients may have the ability to further refine their individual dose schedule according to the inflammatory response to the medication. Once an optimal level of inflammatory response is determined, it may prove possible to interrupt topical application at that point. During the subsequent therapeutic interval, the local skin signs and symptoms may subside while the AKs continue to resolve. REFERENCES 1. Glogau RG. The risk of progression to invasive disease. J Am Acad Dermatol 2000;42(Suppl):S23-s24. 2. Cockerell CJ. Histopathology of incipient intraepidermal squamous cell carcinoma (“actinic keratosis”). J Am Acad Dermatol 2000;42(Suppl):S11-S17. 3. Mittelbronn MA, Mullins DL, Ramos-Caro FA, Flowers FP. Frequency of pre-existing actinic keratosis in cutaneous squamous cell carcinoma. Int J Dermatol 1998;37:677-81. 4. Lober BA, Lober CW. Actinic keratosis is squamous cell carcinoma. South Med J 2000;93:846-50. 5. Goette DK. Topical chemotherapy with 5-fluorouracil. J Am Acad Dermatol 1981;4:633-9.
Salasche, Levine, and Morrison 577
6. Pearlman L. Weekly pulse dosing: effective and comfortable topical 5-fluorouracil treatment of multiple facial actinic keratoses. J Am Acad Dermatol 1991;25:665-7. 7. Unis ME. Short term intensive 5-fluorouracil treatment of actinic keratoses. Dermatol Surg 1995;21:162-3. 8. Dinehart SM. The treatment of actinic keratoses. J Am Acad Dermatol 2000;42:S25-S28. 9. Gupta AK, Weiss JS, Jorizzo JL. 5-fluorouracil 0.5% cream for multiple actinic or solar keratoses of the face and anterior scalp. Skin Therapy Lett 2001;6:1-4. 10. Rivers JK, McLean DI. An open label study to assess the efficacy and safety of topical 3% diclofenac in a 2.5% hyaluronic acid gel for the treatment of actinic keratoses. Arch Dermatol 1997;133:1239-42. 11. Jeffes EW, McCullough JL, Weinstein GD, Kaplan R, Glazer SD, Taylor JR. Photodynamic therapy of actinic keratoses with topical aminolevulinic acid hydrochloride and fluorescent blue light. J Am Acad Dermatol 2001;45:96-104. 12. Kurwas HA, Yong-Gee SA, Seed PT, Markey AC, Barlow RJ. A randomized paired comparison of photodynamic therapy and topical 5-fluorouracil in the treatment of actinic keratosis. J Am Acad Dermatol 1999;41:414-8. 13. Stockfleth E, Meyer T, Benninghoff B, Christopher E. Successful treatment of actinic keratosis with imiquimod cream 5%: a report of a 6 cases. Br J Dermatol 2001;144:1050-8. 14. Persaud AN, Shamuelova E, Sherer D, Lou W, Singer G, Cervera C, et al. Clinical effect of imiquimod 5% cream in the treatment of actinic keratosis. J Am Acad Dermatol 2002;47:553-6. 15. Stockfleth E, Meyer T, Benninghoff B, Salasche S, Papadopoulos L, Ulrich C, et al. A randomized, double blind, vehicle controlled study to assess imiquimod 5% for the treatment of multiple actinic keratoses. Arch Dermatol. In press. 16. Hengge UR, Stark R. Topical imiquimod to treat intraepidermal carcinoma. Arch Dermatol 2001;137:709-11. 17. Mackenzie-Wood A, Kossard S, deLauncey J, Wilkinson B, Owens M. Imiquimod 5% cream in the treatment of Bowen’s disease. J Am Acad Dermatol 2001;44:42-70. 18. Marks R, Gebauer K, Shumack S. Imiquimod 5% cream in the treatment of superficial basal cell carcinoma: results of a multicenter 6-week dose-response trial. J Am Acad Dermatol 2001;44:807-13. 19. Beutner KR, Geisse JK, Helman D, Fox TL, Ginkel A, Owens ML. Therapeutic response of basal cell carcinoma to the immune response modifier imiquimod 5% cream. J Am Acad Dermatol 1991;41:1002-7. 20. Arany I, Tyring SK, Stanley MA, Tomai MA, Miller RL, Smithe MH, et al. Enhancement of the innate and cellular immune response in patients with genital warts treated with topical imiquimod cream, 5%. Antiviral Res 1999;43:55-63. 21. Imbertson LM, Beaurline JM, Couture AM, Gibson JS, Smith RMA, Miller RL, et al. Cytokine induction in hairless mouse and rat skin after topical application of the immune response modifiers, imiquimod and S-28463. J Invest Dermatol 1998;110:734-9. 22. Wagner TL, Ahonen CL, Couture AM, Gibson SJ, Miller RJ, Smith RM, et al. Modulation of TH1 and TH2 cytokine production with the immune response modifiers, R-848 and imiquimod. Cell Immunol 1999;191:10-9. 23. Feldman SR, Fleischer AB, Williford PM, Jorizzo JL. Destructive procedures are the standard of care for treatment of actinic keratoses. J Am Acad Dermatol 1999;40:43-7. 24. Hemmi HT, Kaisho O, Takeuchi S, Sato H, Hoshino T, Horiuchi H, et al. Small anti-viral compounds activate immune cells via the TLR7 MyD88-dependent signaling pathway. Nature Immunology 2002;3:196-200. 25. Epstein E. Does intermittent “pulse” topical 5-fluorouracil therapy allow destruction of actinic keratoses without significant inflammation? J Am Acad Dermatol 1998;38:77-80.
T
here has been a resurgent interest in nonsurgical, topical therapy for actinic keratoses (AKs). Recent publications emphasizing the status of actinic keratosis (AK) as an early stage in the biologic continuum leading to squamous cell carcinoma have kindled this attention.1-4 5-Fluorouracil has been the mainstay therapeutic agent in this area for decades.5-7 However, despite excellent efficacy with this drug, there is a high incidence of adverse effects, which has led to poor patient compliance and reluctance on the part of dermatologists to prescribe it. New formulations of 5-fluorouracil have been developed with the aim of lessening From the University of Arizona Health Sciences Center, Tucson. Funding sources: A grant-in-aid from 3M Pharmaceuticals. Disclosure: Dr Salasche is a medical consultant to 3M Pharmaceuticals. Reprint requests: Stuart J. Salasche, MD, University of Arizona Health Sciences Center, 1501 N Campbell Ave, Tucson, AZ 85724. Copyright © 2002 by the American Academy of Dermatology, Inc. 0190-9622/2002/$35.00 ⫹ 0 16/1/126257 doi:10.1067/mjd.2002.126257
these local skin reactions, and other products have been introduced to eradicate AK.8-12 Early studies with the topical immune response modifier 5% imiquimod have shown promising safety and efficacy profiles for treatment of AK.13-19 The rationale behind the use this new immunomodulatory topical therapeutic agent is that it is believed to stimulate and enhance the local host immune responses against viral infections and skin tumors.20-22 In this open-label study, a novel cycled approach was used in an attempt to maintain high efficacy while lowering the incidence and severity of skin reactions.
METHODS Study design This study was an open-label trial designed to examine the safety and efficacy of topically applied 5% imiquimod cream in the treatment of AKs of the face and scalp by using a cycle approach to therapy. Active medication was applied to the entire treatment area once daily, 3 times a week for 4 weeks. This was followed by a rest period of 4 weeks. The treatment area consisted of an entire cos571
572 Salasche, Levine, and Morrison
metic unit: the scalp, the forehead and temples, or the cheeks. For study qualification, a cosmetic unit was required to contain 5 to 20 discrete, nonhypertrophic AKs. Study participants were allowed to treat two cosmetic units if each satisfied this condition. For qualification in a specific cosmetic area, there could be no overlap in cosmetic units: for example, 3 AKs on the forehead and 2 AKs on the contiguous scalp. The scalp was defined as the area above the superior forehead crease, and the forehead-temple area extended below this crease to the eyebrow. Laterally, the forehead-temple area extended to the hairline above a line drawn from the lateral canthus to the superior attachment of the ear. The cosmetic junction lines of the lower eyelid, nose, lips, jaw line, and ear bound the cheeks. Study subjects were excluded if they had, within the prior month, used any topical medication within the treatment area that may have efficacy against AK such as topical retinoids, hydroxy-acids, 5-fluorouracil, or diclofenac or if they had received photodynamic therapy. Subjects were similarly excluded if they had had a dermabrasion, chemical peel, or laser resurfacing in the treatment area within 6 months. Immunosuppressed patients were excluded. The primary efficacy end point was complete clearance of AKs within the treatment area. No attempt was made to discriminate between AKs noted at baseline examination and AKs that arose in the intervening skin (subclinical AK) during treatment. If any AKs were present at the end a 4-week rest period, the entire area was re-treated for another cycle of 4 weeks, followed by another 4-week rest period. Each cycle then consisted of 8 weeks. A maximum of 3 cycles was permitted. If AKs were present after the third cycle (24 weeks), this was considered a treatment failure. The secondary efficacy end point was a 75% reduction in AKs over baseline. This secondary end point would correspond to a clinically relevant outcome. Histologic examination was not performed. As in the clinical situation, AK counts before, during, and after treatment were done by visual inspection and palpation. A lesion was considered an AK if the investigator would have ordinarily treated any individual lesion by destruction with liquid nitrogen. If the subject experienced moderate to severe local skin reactions during the active treatment part of the cycle, an early rest period would be initiated. The decision to impose a rest period was also influenced by the presence of symptoms that included pain, itching, burning, or tenderness. A rest period would also be imposed if systemic adverse events occurred. At the end of this rest period, a decision about whether to resume treatment would be made. This depended on
J AM ACAD DERMATOL OCTOBER 2002
the nature of the inciting event and/or whether AKs remained at the end of the rest period. The intent was to treat until there was complete clearing or a treatment failure at the end of 3 cycles. Participants were asked to apply the cream evenly over the entire cosmetic unit. Because one sachet covers approximately 20 to 30 cm2, patients were asked to use one sachet for the cheek, the forehead, and the anterior bald scalp. If the entire scalp were to be treated, two sachets would be required. Safety Study subjects were seen at baseline and then every 2 weeks until the study was complete, that is, either complete clearance of AKs or the end of 3 cycles (24 weeks). At each visit the following were done: a complete AK count, digital photography of the treatment area, evaluation of adverse local or systemic events, and an examination of the treatment site to assess local skin reactions. Local skin reactions were graded as mild, moderate, or severe. The signs that were assessed included erythema, edema, induration, erosion, ulceration, scabbing/ crusting, exudation, vesicles, and flaking/dryness/ scale. Symptoms of pain, burning, or itching were also quantified by using the same scale. Participants were questioned at each visit about any change in their health. They were specifically asked whether they had experienced any flu-like symptoms to include fever, chills, muscle fatigue, or a general feeling of weakness. Study population Twenty-five study subjects were enrolled and gave informed consent to the institutional review board–approved trial protocol. Thirty-three cosmetic unit sites on these 25 patients were included in the trial. Eight subjects qualified for 2 sites and chose to treat both areas. There were 23 men and 2 women. Anatomic site distribution was as follows: scalp, 13; forehead, 13; and cheek, 7. The average age of the study population was 73 years. This original group of 25 subjects and 33 sites constituted the basis of our intent-to-treat analysis, even though several subjects left the study for nonmedical reasons or did not comply with a sufficient number of doses to have qualified. Two subjects moved away from Tucson in mid-trial for personal reasons and were considered to have experienced treatment failures for intent-to-treat analysis purposes. Another subject withdrew consent after only 3 treatments because he was blind and his wife did not want to apply the cream to his scalp. This case was also considered a treatment failure.
J AM ACAD DERMATOL VOLUME 47, NUMBER 4
Salasche, Levine, and Morrison 573
Fig 1. A, Baseline treatment of forehead 3 times a week; actinic keratosis (AK) count ⫽ 19. B, At 2 weeks; AK count ⫽ 35. C, At 4 weeks; AK count ⫽ 40. D, At 6 weeks; AK count ⫽ 1. E, At 8 weeks; AK count ⫽ 0.
RESULTS Efficacy The primary efficacy outcome was 100% clearing of the AKs within the treatment area. On the basis of this criterion, there was a high clearance rate. For the intent-to-treat analysis, 82% (27/33) of the sites were
completely cleared (Fig 1). The 6 treatment failures included sites in two subjects who moved away in mid-trial and the one who withdrew consent after only 3 medication applications. Two other study subjects had accelerated brisk inflammatory reactions that necessitated an early rest period. In each
574 Salasche, Levine, and Morrison
J AM ACAD DERMATOL OCTOBER 2002
Table I. Total clearance by cycle: intent to treat sites (n ⫽ 33) No. cleared
%
15 12 0 27
46 36 0 82
Cycle 1 Cycle 2 Cycle 3 Totals
Table III. Total and average actinic keratosis counts from all 33 sites during first 8-week cycle Baseline Week 2 Week 4 Week 6 Week 8
No. of AKs
Mean
368 613 515 228 144
12 19 16 7 5
AKs, Actinic keratosis.
Table II. Time to complete clearance Cycle
Clear during treatment period Clear at 2nd wk of rest Clear at 4th wk of rest Total
First
Second
Third
2 2 11 15
2 2 8 12
0 0 0 0
of these cases, two sites were being treated. In both instances, at the end of the rest period, one anatomic site was clear and the other was not. Both subjects declined to resume treatment of the involved site, and each site was considered a treatment failure. A single subject went through three 8-week cycles and had persistent AK lesions. Among the cases of total clearance failures, one subject, the only one who completed the entire 3-cycle course, approached the 75% clearance required for the secondary efficacy end point. However, only 73% clearance was achieved because 3 AKs remained from a baseline count of 11. All subjects who experienced complete clearance did so within the first 2 cycles (Table I). Most experienced clearance during the rest periods (Table II). Table III summarizes the serial AK count data. Because complete AK counts of the treatment area were recorded at 2-week intervals, a time course of targetlesion response could be tabulated. The baseline AK average score of 12 AKs was reduced to an average of 5 AKs by the end of the first full cycle (week 8). The observation that AK counts increased in the weeks after initiation of treatment indicates that incipient, subclinical AKs were being uncovered and treated. The continued reduction in number of AKs during the rest period defines the concept of the “therapeutic interval.” As can be seen from Table IV, several study subjects who still had AKs at the end of 2 weeks into the rest period experienced complete clearance by the end of another 2 weeks without medication. Safety Overall, the dosing regimen was well tolerated by the majority of patients. Most of the local skin reactions were of the mild to moderate category. These
Table IV. Value of 4-week rest period Subjects with clearance during last 2 wk of rest who were positive after first 2 wk First cycle
9
Second cycle
8
tended to begin in the second to third week of active treatment and persisted through the fourth week when the rest period began. Resolution began promptly after cessation of active treatment. Severe reactions occurred in 5 patients, 4 of whom required early rest periods (Table V, Fig 2). All severe reactions occurred during the first cycle in the setting of an early, intense inflammatory reaction. In each case an accelerated response was noted within 1 to 2 weeks of initiating therapy. Erythema of the lesion and surrounding skin predominated and was accompanied by intense edema, induration, and enlargement of the AKs. This was followed by erosion and crusting of the AKs, leading to ulceration if the medication was not discontinued. Sensations of burning, pain, and tenderness accompanied these severe intense inflammatory reactions.
DISCUSSION A common approach to the treatment of patients with AK is to destroy small numbers of scattered, discrete AKs with liquid nitrogen.8,23 A different situation exists when there are numerous AKs in any given area or multiple areas of the exposed head and neck region and extremities. These may be best managed with a topical medication that eradicates not only the visible lesions but also the subclinical nascent lesions. Imiquimod is an imidazoquinolinamine, one of the new immune response modifiers. It has been approved by the Food and Drug Administration for the treatment of anogenital warts but has been studied as off-label treatment for several other skin indications including AK, basal cell carcinoma, and Bo-
Salasche, Levine, and Morrison 575
J AM ACAD DERMATOL VOLUME 47, NUMBER 4
Fig 2. A, Baseline; actinic keratosis (AK) count ⫽ 15. B, At 3 weeks: edema, erythema, crusting, tenderness; treatment discontinued; AK count ⫽ 30. C, At 4 weeks: resolution of adverse signs and symptoms; residual erythema. D, At 9 weeks: complete resolution of AKs and inflammation.
Table V. Summary of adverse events leading to early rest periods Patient No./Sex
11/F 21/M 23/F 24/M
Area
Forehead; cheek Forehead Forehead; cheek Forehead
Rest begun
3 wk 2 wk 2 wk 2 wk
Symptoms
Skin burning, severe erythema and erosions, fever, chills Mod erosions, erythema, burning Systemic malaise, aches; severe erythema, exudative crusts Burning, severe erythema
Skin type
II/Severe II/Severe I/Severe II/Severe
Mod, Moderate.
wen’s disease.13-19 The exact mechanism of action is not known but is likely related to the findings of preclinical and clinical research showing stimulation of both innate and acquired immunity. Monocytes, macrophages, and dendritic cells are induced to secrete cytokines that initiate a favorable inflammatory cascade, resulting in tumor necrosis. Cytokines interferon alfa, interleukin 12, and tumor necrosis factor ␣ are all generated locally. These in turn
activate natural killer cells of the innate immune system and induce the production of interferon ␥ that upregulates cell-mediated immunity.20-22 Recent data indicate that immune response modifiers act through TLR7 receptor (Toll-like receptor 7).24 Observations from this investigation with topical imiquimod indicate that when it is applied 3 times a week, all the AKs in the treatment area respond in a synchronized manner. Individual lesions become er-
576 Salasche, Levine, and Morrison
ythematous, edematous, and indurated. Subsequently, they become centrally eroded and crusted, and over the ensuing weeks, become smaller and ultimately disappear. In an earlier study, the peak inflammatory reaction took place between 4 and 6 weeks of treatment when a 3-times-a-week continuous, uninterrupted dosing regimen was used.15 The purpose of this open-label trial was to test the hypothesis that topically applied 5% imiquimod cream could be administered in a cycled fashion to attain the following results: (1) initiation of a therapeutic inflammatory response, (2) interruption of dosing at a time before the inflammatory reaction reaches a magnitude that would interfere or restrict daily activities, and (3) continued efficacy during the rest period as inflammation subsides. This preliminary open-label trial indicates that such a cycled approach of alternating 4 weeks of active treatment with rest periods of equal length may achieve high efficacy while minimizing adverse effects. This leads to high rates of patient acceptance, compliance, and satisfaction. This regimen is not to be confused with the “pulsed” therapy for AK with topical 5-fluorouracil in which topical application is done twice a day, one day a week until lesions are gone.6,7,25 Therapeutic interval: The primary benefit of cycled therapy Because we adhered to a rather rigid 4-week active intervention phase followed by a 4-week imposed rest period, we identified a “therapeutic interval.” During this gap (the rest period) between cessation of topical application of 5% imiquimod and restarting the treatment if any AKs remained, inflammation steadily diminished while the therapeutic effect continued. Treatment beyond the 4-week period probably would have, in many instances, resulted in unacceptable inflammation, unsightly appearance, and unscheduled cessation of treatment. Although inflammation had subsided sufficiently by 2 weeks into the rest period in most cases to have allowed the patient to restart topical application, there was continued improvement and clearing of AKs during the subsequent 2 weeks of the rest period (Tables II and IV). In the first cycle, 9 subjects with AKs who were 2 weeks into the rest period experienced complete clearance at the end of the 4-week rest period. In the second cycle, the same occurred with 8 additional subjects. Until sequential “mechanism of action” studies are done, we can only speculate that one of two possible events may explain these observations. It is possible that AK destruction is due to some critical molecular event that occurs early in the treatment cycle that programs the AKs for eradication. What
J AM ACAD DERMATOL OCTOBER 2002
we noted clinically is only the visible evidence of lesion regression. Alternatively, there may be an ongoing or self-sustaining therapeutic effect from the cytokine cascade initiated by the imiquimod. Although a 4-week rest period works well, it is still not clear whether there is an optimal active treatment time. For most patients in this trial, the 4-week, 3-times-a-week regimen was effective. A subgroup reacted very quickly and intensely, usually by 2 weeks. Although most of these rapid responders ultimately experienced clearance, they probably would have been much more comfortable with a slower, less intense response. We have identified potential risk factors for a local reaction, which include Fitzpatrick type I or II (severe to very severe) photodamage in the treatment area, many AKs within the treatment area, and female sex. Perhaps those with these risk factors should have a regimen of less frequent dosing (1-2 times per week), a shorter duration of active treatment (1-3 weeks), or both. This will be tested in future prospective, randomized trials. Furthermore, it may prove possible to titrate dosing during active treatment. When a predetermined level of inflammation or erosion/ crusting is attained, active treatment could be halted and the therapeutic interval could be initiated. In the clinical setting, complete clearance would not necessarily be the relevant end point. If only a few AKs remained at the end of a rest period, one might choose not to re-treat the entire area with subsequent cycles. Rather, focal lesions could be destroyed with cryotherapy or the patient could “spot” treat the remaining lesions with imiquimod cream. Subclinical AK By the time AKs develop, the skin in which they appear has been subjected to years of excessive sun exposure and photodamage, which results in varying degrees of hypopigmentation and hyperpigmentation, dermal thickening caused by solar elastosis, epidermal atrophy, and telangiectasia. In addition to the fully formed and clinically detectable AKs, the skin also manifests many undetectable clonally expanded foci of dyskeratotic keratinocytes. Although many of these “AKs-to-be” are probably destroyed by ongoing immune surveillance and local immune reactions, some eventually manifest as clinical AKs. One of the important salutary benefits of topical imiquimod treatment over cryotherapy of focal lesions is that not only are clinically evident AKs eradicated, but the surrounding photodamaged skin is also treated at the same time. In this study, the AK counts were noted to be elevated over baseline during the first cycle in almost every subject (Table III). Because the baseline AKs were merely counted
J AM ACAD DERMATOL VOLUME 47, NUMBER 4
and not sequentially tracked with a template during the study, it was impossible to discern the baseline lesions from the newly “unmasked” AKs. By the time the subclinical AKs were noted (by inference from an increased count over baseline), all the AKs within the treatment area were synchronized to the same stage of inflammation or resolution and had a similar clinical appearance. Finally, practicing clinicians should be cautioned not to construe the high complete clearance rates attained in this trial as the last word on this subject. It must be stressed again that this is an open-label pilot study whose results need to be corroborated by larger, prospective, randomized controlled studies. Such studies have been initiated. In a similar vein, it has not be ascertained whether the areas cleared by topical imiquimod in this clinical trial remain free of disease or whether new AKs will recur in the treated areas, and if so, when they will appear. To help answer these questions, the study group from this trial will participate in a follow-up study. The plan is to examine each patient every 3 months for 2 years. The time to appearance of any new AKs and the number of new AKs will be determined and subsequently reported. Conclusions The findings of this study must be evaluated cautiously. It is an open-label trial in a small number of study subjects. Safety, efficacy, and duration of efficacy must be corroborated by larger, randomized studies. However, cycle therapy and identification of a therapeutic interval may be a first step in allowing dermatologists to tailor dosing regimens to the specific patient characteristics according to the degree of sun damage and skin type. Furthermore, patients may have the ability to further refine their individual dose schedule according to the inflammatory response to the medication. Once an optimal level of inflammatory response is determined, it may prove possible to interrupt topical application at that point. During the subsequent therapeutic interval, the local skin signs and symptoms may subside while the AKs continue to resolve. REFERENCES 1. Glogau RG. The risk of progression to invasive disease. J Am Acad Dermatol 2000;42(Suppl):S23-s24. 2. Cockerell CJ. Histopathology of incipient intraepidermal squamous cell carcinoma (“actinic keratosis”). J Am Acad Dermatol 2000;42(Suppl):S11-S17. 3. Mittelbronn MA, Mullins DL, Ramos-Caro FA, Flowers FP. Frequency of pre-existing actinic keratosis in cutaneous squamous cell carcinoma. Int J Dermatol 1998;37:677-81. 4. Lober BA, Lober CW. Actinic keratosis is squamous cell carcinoma. South Med J 2000;93:846-50. 5. Goette DK. Topical chemotherapy with 5-fluorouracil. J Am Acad Dermatol 1981;4:633-9.
Salasche, Levine, and Morrison 577
6. Pearlman L. Weekly pulse dosing: effective and comfortable topical 5-fluorouracil treatment of multiple facial actinic keratoses. J Am Acad Dermatol 1991;25:665-7. 7. Unis ME. Short term intensive 5-fluorouracil treatment of actinic keratoses. Dermatol Surg 1995;21:162-3. 8. Dinehart SM. The treatment of actinic keratoses. J Am Acad Dermatol 2000;42:S25-S28. 9. Gupta AK, Weiss JS, Jorizzo JL. 5-fluorouracil 0.5% cream for multiple actinic or solar keratoses of the face and anterior scalp. Skin Therapy Lett 2001;6:1-4. 10. Rivers JK, McLean DI. An open label study to assess the efficacy and safety of topical 3% diclofenac in a 2.5% hyaluronic acid gel for the treatment of actinic keratoses. Arch Dermatol 1997;133:1239-42. 11. Jeffes EW, McCullough JL, Weinstein GD, Kaplan R, Glazer SD, Taylor JR. Photodynamic therapy of actinic keratoses with topical aminolevulinic acid hydrochloride and fluorescent blue light. J Am Acad Dermatol 2001;45:96-104. 12. Kurwas HA, Yong-Gee SA, Seed PT, Markey AC, Barlow RJ. A randomized paired comparison of photodynamic therapy and topical 5-fluorouracil in the treatment of actinic keratosis. J Am Acad Dermatol 1999;41:414-8. 13. Stockfleth E, Meyer T, Benninghoff B, Christopher E. Successful treatment of actinic keratosis with imiquimod cream 5%: a report of a 6 cases. Br J Dermatol 2001;144:1050-8. 14. Persaud AN, Shamuelova E, Sherer D, Lou W, Singer G, Cervera C, et al. Clinical effect of imiquimod 5% cream in the treatment of actinic keratosis. J Am Acad Dermatol 2002;47:553-6. 15. Stockfleth E, Meyer T, Benninghoff B, Salasche S, Papadopoulos L, Ulrich C, et al. A randomized, double blind, vehicle controlled study to assess imiquimod 5% for the treatment of multiple actinic keratoses. Arch Dermatol. In press. 16. Hengge UR, Stark R. Topical imiquimod to treat intraepidermal carcinoma. Arch Dermatol 2001;137:709-11. 17. Mackenzie-Wood A, Kossard S, deLauncey J, Wilkinson B, Owens M. Imiquimod 5% cream in the treatment of Bowen’s disease. J Am Acad Dermatol 2001;44:42-70. 18. Marks R, Gebauer K, Shumack S. Imiquimod 5% cream in the treatment of superficial basal cell carcinoma: results of a multicenter 6-week dose-response trial. J Am Acad Dermatol 2001;44:807-13. 19. Beutner KR, Geisse JK, Helman D, Fox TL, Ginkel A, Owens ML. Therapeutic response of basal cell carcinoma to the immune response modifier imiquimod 5% cream. J Am Acad Dermatol 1991;41:1002-7. 20. Arany I, Tyring SK, Stanley MA, Tomai MA, Miller RL, Smithe MH, et al. Enhancement of the innate and cellular immune response in patients with genital warts treated with topical imiquimod cream, 5%. Antiviral Res 1999;43:55-63. 21. Imbertson LM, Beaurline JM, Couture AM, Gibson JS, Smith RMA, Miller RL, et al. Cytokine induction in hairless mouse and rat skin after topical application of the immune response modifiers, imiquimod and S-28463. J Invest Dermatol 1998;110:734-9. 22. Wagner TL, Ahonen CL, Couture AM, Gibson SJ, Miller RJ, Smith RM, et al. Modulation of TH1 and TH2 cytokine production with the immune response modifiers, R-848 and imiquimod. Cell Immunol 1999;191:10-9. 23. Feldman SR, Fleischer AB, Williford PM, Jorizzo JL. Destructive procedures are the standard of care for treatment of actinic keratoses. J Am Acad Dermatol 1999;40:43-7. 24. Hemmi HT, Kaisho O, Takeuchi S, Sato H, Hoshino T, Horiuchi H, et al. Small anti-viral compounds activate immune cells via the TLR7 MyD88-dependent signaling pathway. Nature Immunology 2002;3:196-200. 25. Epstein E. Does intermittent “pulse” topical 5-fluorouracil therapy allow destruction of actinic keratoses without significant inflammation? J Am Acad Dermatol 1998;38:77-80.