Imiquimod 5% cream for the treatment of superﬁcial basal ... 5%...Imiquimod 5% cream for the treatment of superﬁcial basal cell carcinoma: A double-blind, randomized, vehicle-controlled

Imiquimod 5% cream for the treatment ofsuperficial basal cell carcinoma: A double-blind,

randomized, vehicle-controlled study*

John K. Geisse, MD,a Phoebe Rich, MD,b Amit Pandya, MD,c Kenneth Gross, MD,d

Kara Andres, MS,e Angie Ginkel, BS,e and Mary Owens, MDe Vallejo and San Diego, California;Portland, Oregon; Dallas, Texas; and St Paul, Minnesota

Background: Imiquimod 5% cream may provide an effective nonsurgical treatment for superficial basalcell carcinoma (sBCC) based on results of previous studies.

Objective: The objective of this phase II dose-response study was to explore various dosing regimens usingimiquimod 5% cream for sBCC to find the most effective frequency of dosing with tolerable side effects.

Methods: Patients (n � 128) were dosed twice daily, once daily, 5 times a week, or 3 times a week in this12-week, randomized, double-blind, vehicle-controlled study. At 6 weeks after treatment, the entire tumorarea was clinically evaluated, excised, and examined exhaustively for histologic evidence of residual sBCC.

Results: Complete response rates were 100% (10/10), 87.1% (27/31), 80.8% (21/26), and 51.7% (15/29) forpatients in the twice daily, once daily, 5 times a week, and 3 times a week imiquimod groups, respectively,and 18.8% (6/32) in the vehicle group.

Conclusion: Imiquimod 5% cream was effective in the treatment of sBCC. Daily or 5 times a week dosingfor 12 weeks demonstrated high efficacy results with acceptable safety profiles. (J Am Acad Dermatol 2002;47:390-8.)

B asal cell carcinoma (BCC) is the most com-mon cancer in the United States, with ap-proximately 1 million new cases estimated

annually. An increasing incidence has been ob-served in the past several years in the United Statesand other parts of the world.1,2 There are several

histologic subtypes of BCC with significant differ-ences in site distribution and clinical outcome be-tween subtypes.3-5 Although nodular BCC (nBCC) isthe dominant subtype for all anatomic sites, super-ficial BCCs (sBCCs) are commonly seen on the trunkand limbs, and appear at an earlier age.6

Surgical excision and curettage-electrodesicca-tion are the most common treatments for primary,nonaggressive BCC. These procedures are effectiveand have low recurrence rates when used to treatprimary BCC on low-risk anatomic sites (neck,trunk, and extremities).7,8 A total of 36% of patientswho have 1 BCC have additional primary BCCs, andmany of these patients experience multiple newBCCs.9 Imiquimod may offer an alternative to re-peated surgical or ablative treatments.

Imiquimod cream, 5%, a drug approved in theUnited States under the trade name Aldara for thetopical treatment of external genital/perianal warts,is a novel immune response modifier that inducesinterferon production by monocytes/macrophages.Imiquimod has induced immunologic activity in hu-man skin infected with human papillomavirus(HPV), stimulating significant increases in mRNA forinterferon-� (IFN-�), 2�5�-oligoadenylate synthetase(2�5�-AS) and IFN-�.10 Increases in mRNA for CD4,

From Solano Dermatology Associates, Vallejoa; Northwest Cutane-ous Research Specialists, Portlandb; University of Texas South-western Medical Center, Dallasc; Skin Surgery Medical Group, Inc,San Diegod; and 3M Pharmaceuticals, St Paul.e

*A list of all enrolling investigators can be found at the end of thearticle.

Sponsored by 3M Pharmaceuticals, St. Paul, Minn. Drs Geisse, Rich,Pandya, and Gross have received support from 3M Pharmaceuti-cals for performing clinical trials. Drs Geisse and Rich have servedas consultants for 3M Pharmaceuticals. Drs Pandya and Grosshave received honoraria for participation in a speaker’s bureau.Dr Owens and Ms Andres are employees of 3M Pharmaceuticals,and Ms Ginkel is a former employee.

An abstract and poster based on this study were presented at the8th World Congress on Cancers of the Skin, Zurich, Switzerland,July 18-21, 2001.

Accepted for publication February 19, 2002.Reprint requests: John K. Geisse, MD, Solano Dermatology Associ-

ates, 127 Hospital Dr, Suite 204, Vallejo, CA 94589.Copyright © 2002 by the American Academy of Dermatology, Inc.0190-9622/2002/$35.00 � 0 16/1/126215doi:10.1067/mjd.2002.126215

390

CD8, and tumor necrosis factor-� (TNF-�) were alsoobserved, suggesting activation of a helper T-celltype 1-mediated response.10 IFN alone has signifi-cant activity against BCC, but requires multiple in-tralesional injections that can be difficult to admin-ister, and the treatment can result in unpleasant butusually transient side effects.11-13

Imiquimod has a unique mechanism of action, and,because it is applied by the patient, imiquimod 5%cream may be more convenient and well tolerated forBCC than many current treatments. In a pilot study,imiquimod 5% cream demonstrated clinical efficacy inthe treatment of BCC.14 The objective of the presentstudy was to establish a safe and effective treatmentregimen using topical imiquimod 5% cream in a largerseries of patients for the treatment of sBCC.

PATIENTS AND METHODSPatients

Male or female patients, 18 years of age or older,who had a histologically confirmed diagnosis ofsBCC, were eligible for study participation. The tar-get tumor site excluded areas within 1 cm of thehairline, eyes, nose, mouth, or ears; the anogenitalarea; and hands and feet. The target tumors wereprimary, noninfected, measured between 0.5 and2.0 cm2, and could not have been previously treated,recurrent, or within 5 cm of another BCC tumor.

MethodsThis phase II, randomized, double-blind, vehicle-

controlled, dose-response study was conducted in13 study centers in the United States, consisting of apretreatment screening visit, a 12-week treatmentperiod with regularly scheduled interval visits, and a6-week posttreatment visit for surgical excision ofthe target tumor area. At the screening visit, patientswere informed of study procedures and of theirrights and responsibilities as study participants; allpatients signed an institutional review board ap-proved, written, informed consent document.

A biopsy specimen of no more than 25% of thetumor area was taken for histologic confirmation ofsBCC. Biopsy specimens were processed at one cen-ter but read by 2 independent, blinded dermato-pathologists. Superficial BCC was defined as tumorconfined in the specimen to the superficial reticularand papillary dermis arranged as small tumor nestsat the undersurface of the epidermis.

If a BCC contained nodular or micronodular com-ponents, or if the basaloid cells extended deep intothe reticular dermis, the tumor was not classified assBCC and was not eligible for the study. A consensusof the 2 dermatopathologists’ histologic diagnoseswas required for patient enrollment.

A baseline area of the target tumor was deter-

mined at the treatment initiation visit by measuringand multiplying the 2 largest perpendicular dimen-sions of the target tumor. The target tumor site andappropriate surrounding anatomic landmarks weremapped using a clear, flexible polypropylene sheetas a template so that the correct site could be foundfor excision at the end of the study.

Treatment. Patients were randomized to receiveeither imiquimod or vehicle in 1 of 4 dosing regi-mens: twice daily, once daily, 5 times a week, or 3times a week according to a computer-generatedrandomization schedule. Upon randomization, pa-tients were assigned a unique identification numberin numerical sequence. Imiquimod 5% cream andvehicle cream were in identical packaging as 250 mgsingle-use sachets. Both patients and investigatorswere blinded to the content of the treatment cream(active or vehicle) dispensed.

Patients applied imiquimod 5% or vehicle cream tothe target tumor until 12 weeks of treatment werecompleted. Patients were prescribed rest periods asneeded up to a total of 14 days, for local skin reactions,or for treatment site adverse events. Rest periods werecounted as part of the 12-week treatment period.

The estimated amount of study cream applied ateach dose was determined by the larger diameter ofthe target tumor. Tumors up to 1 cm2 in size weredosed with a cream droplet approximately 4 mm indiameter (0.5 mg imiquimod). Tumors ranging from1 cm2 to �1.5 cm2 were covered with 5 mm (1.25 mgimiquimod) and tumors �1.5 cm2 were coveredwith 7 mm (2 mg imiquimod).

Safety evaluations. To assess safety, patientsreturned to the clinic for designated visits for eval-uation, photographs and documentation of tumorappearance, local skin reactions, vital sign measure-ments, adverse events, and concomitant medica-tions.

Local skin reactions. A visual assessment of thetarget tumor area was performed by the investigatorat each clinic visit. Local skin reaction categorieswere erythema, edema, induration, vesicles, ero-sion, ulceration, excoriation/flaking, and scabbing.A 4-point scale was used where: 0 � none, 1 � mild,2 � moderate, 3 � severe.

Patient-reported signs or symptoms correspond-ing to the above local skin reactions and other ad-verse events were recorded separately at each visit.

Efficacy evaluations. At the 6-week posttreat-ment visit, the investigator made a clinical assess-ment to determine whether BCC was visible at thetarget tumor site. All patients then had the entiretarget tumor area and a 3-mm margin surgicallyexcised and histopathologically examined for evi-dence of residual tumor. A patient was considered a

Geisse et al 391J AM ACAD DERMATOL

VOLUME 47, NUMBER 3

complete responder if there was no histologic evi-dence of BCC in the excised posttreatment targettumor tissue.

The excision specimen was bread-loaf sectionedfrom tip to tip in 2- to 4-mm thick slices, which werethen paraffin embedded. The specimens were step-sectioned at least every millimeter in the center of thespecimen and at least every 3 mm through the tips.Sections were stained with hematoxylin and eosin andhistologically examined for residual BCC using lightmicroscopy. All sections were examined by two inde-pendent, blinded dermatopathologists for evidence ofany residual BCC and to ensure that the excisionalmargins were free of tumor. At least 6 and up to 20 ormore slides were examined per case, depending uponthe size of the ellipse that was excised.

Statistical evaluations. Sample size calculationswere based on the ability to detect differences incomplete response rates. Eighteen patients in eachof the 3 imiquimod dose groups and 18 patients inthe combined vehicle group gave this study at least80% power to detect a difference in complete re-sponse rates of 20% for vehicle patients versus 80%for any of the imiquimod dose groups, assuming thateach imiquimod-versus-vehicle comparison was car-ried out at an alpha level of 0.05/3. The statisticalanalysis was based on the intent-to-treat data set,which included all randomized patients. The pri-mary variable was the complete response rate, de-fined as the proportion of patients who had nohistologic evidence of BCC in the posttreatment ex-cision specimen from the target tumor area. The datafrom all the vehicle dosing groups were pooled todetermine the combined vehicle response rate. Fish-er’s exact tests were used to compare the completeresponse rates of the once-daily, 5-times-a-week,and 3-times-a-week imiquimod groups to the com-

bined vehicle group in a pairwise manner. Each testwas carried out at an alpha level of 0.05/3; this Bon-ferroni adjustment was made to preserve the overallalpha level at 0.05. The response rate of the twice-dailyimiquimod group was not compared to vehicle be-cause of the small sample size for this group.

A secondary analysis was completed to comparethe investigator’s clinical assessment of target tumorclearance at the 6-week posttreatment visit to thehistologic evaluation.

RESULTSA total of 128 patients, including 82 men and 46

women, were enrolled and randomized beginning July22, 1998. The mean age was 59 years (range, 35 to 85years). All patients were white with Fitzpatrick skintypes I-IV: 13% had skin type I, 46% had skin type II,34% had skin type III, and 7% had skin type IV.15

Of the 128 patients, 92 (71.9%) had undergone atleast 1 previous excision for BCC and 68 (53.1%) hadnontarget BCC tumors present at study initiation(Table I). A total of 96 patients were randomized toimiquimod treatment groups and 32 patients to thevehicle treatment group. Although 10 patients wererandomized to the twice-daily regimen, no addi-tional patients were enrolled into this treatmentgroup because of the incidence and severity of re-ported local reactions in this regimen.

Of the 128 randomized patients, 24 (18.8%) with-drew from the treatment portion of the study. Over-all, 13 (10.2%) patients discontinued treatment be-cause of local skin reactions, including 3 from thetwice-daily group, 7 from the once-daily group, and3 from the 5-times-a-week group. Four (3.1%) pa-tients discontinued because of application site ad-verse events, including 1 from the twice-daily group,2 from the once-daily group and 1 from the 5-times-

Table I. Patient demographics and target tumor characteristics

Demographics

Combinedvehiclen � 32

Imiquimod

All patientsn � 128

3 times a weekn � 29


Once dailyn � 31

Twice dailyn � 10

GenderFemale 12 13 8 11 2 46Male 20 16 18 20 8 82

Age — mean years (range) 58 (38-85) 62 (36-85) 55 (38-84) 56 (35-85) 69 (51-85) 59 (35-85)Patients with history of BCC excision 24 20 19 21 8 92Patients with nontarget tumors 15 18 12 19 4 68Target tumor location

Neck/face/forehead 3 1 1 2 1 8Upper extremity (not hand) 11 7 6 9 4 37Trunk: anterior 7 7 11 12 2 39Trunk: posterior 8 12 5 6 2 33Lower extremity/thigh (not foot) 3 2 3 2 1 11

Median target tumor size 0.8 cm2 1.0 cm2 0.6 cm2 0.7 cm2 1.0 cm2 0.8 cm2

392 Geisse et al J AM ACAD DERMATOL

SEPTEMBER 2002

a-week group. Some of these discontinuations werenot the choice of the patient, but were imposed bythe investigator because the maximum 14-day restfrom dosing allowed by the protocol had been ex-ceeded. Also, 2 (1.6%) discontinued because of non-compliance, 2 (1.6%) for personal reasons, 2 (1.6%)were lost to follow-up, and 1 (0.8%) was discontin-ued for a laboratory abnormality identified at initia-tion. Despite the fact that nearly 20% of patientsdiscontinued treatment before the end of the 12-week treatment period, only 3 (2%) patients werenot followed to completion. Of the 128 patientsrandomized, 125 completed the end of study proce-dures and histologic evaluations by January 19,2000.

Target tumor characteristicsOut of 128 patients, 116 (90.6%) had biopsy-

confirmed prestudy diagnoses of sBCC. Twelve pa-tients had confirmed prestudy histologic diagnosesof nBCC after being randomized with a preliminarydiagnosis of sBCC; if even a small nodular compo-nent was evident histologically, the biopsy was clas-sified as nBCC for study purposes. These 12 patientswere included in the analysis. The median targettumor size after biopsy at the initiation visit for eachtreatment group is presented in Table I. Locations ofthe target tumors were primarily on the upper body.The most frequently reported target tumor locationwas upper extremities (excluding the hand), re-ported by 37 (28.9%) of 128 patients.

Amount of imiquimod appliedThere was substantial variability in the amount of

study drug applied from week to week in eachimiquimod dose group because of patient discon-tinuations and rest periods. A total of 41 patientstook rests from study cream application during thetreatment period. In general, a decreasing percent-age of patients took rests as the dosing frequency ofimiquimod decreased and the median number ofdoses rested was lower with less frequent dosing(Table II). No vehicle patients took rests. Over the12-week course of treatment, the mean totalamounts of imiquimod applied were 146 mg for thetwice-daily group, 69 mg for the once-daily group,

and 43 mg for both the 5-times-a-week and the3-times-a-week groups.

Treatment responseOf the 128 patients included in the intent-to-treat

data set, 125 (97.7%) had posttreatment excisions.Depending on the size of the tumor treated, anaverage of 125 histologic sections were examinedfor each excision specimen, with over 100 sectionsfor most. The 3 patients who did not complete theposttreatment excision were lost to follow-up (oneon vehicle and one on 5 times a week imiquimoddosing) or noncompliant (one on once daily imi-quimod dosing). These 3 were counted as treatmentfailures in the statistical analysis. There was a posi-tive association between dosing frequency and com-plete response rate; higher response rates were as-sociated with more frequent dosing (Fig 1). Fisher’sexact tests comparing the 3 imiquimod dosinggroups (once daily, 5 times a week, and 3 times aweek) to the vehicle group in a pairwise mannerfound statistically significant differences betweengroups with respect to complete response rates (allP values � .05/3) (Figs 2-6).

Although the number of patients with mixed su-perficial and nodular histology enrolled into eachtreatment group was too small to analyze separately,9 of the 12 patients (including all 6 of those treatedonce daily or 5 times a week) were complete re-sponders to treatment. Similarly, the number of pa-tients enrolled into the twice-daily, once-daily, and5-times-a-week treatment groups who discontinuedtreatment before completing 12 weeks of dosingwere small and not analyzed separately for efficacy.However, 5 of 5 patients treated twice daily, 8 of 10treated once daily, and 4 of 6 treated 5 times a weekwho discontinued treatment early were completeresponders. For these responders who discontinuedearly, the length of treatment varied, ranging from 4to 11 weeks for the once-daily and 5-times-a-weektreatment groups.

Investigator assessment versus excisionresults

The ability of the investigators to correctly assess thetarget site for tumor evidence was analyzed by com-

Table II. Rest periods taken by patients in the imiquimod dosing groups

Rest period data

Imiquimod



Once dailyn � 31

Twice dailyn � 10

Patients taking rest periods from dosing 6 (21%) 7 (27%) 21 (68%) 7 (70%)Median number of doses rested (range) 3 (3-6) 10 (5-11) 13 (1-19) 14 (11-30)First week that rests were taken: median (range) Week 7 (3-9) Week 5 (2-9) Week 5 (2-9) Week 5 (2-7)


VOLUME 47, NUMBER 3

paring the investigator assessment of the target tumor siteat the 6-week posttreatment excisional visit with the his-tologic results (Table III). For all imiquimod groups com-bined, the negative predictive value for the investigatorassessment was 100% (38/38). Therefore, all patients clin-ically assessed as negative (no target tumor visually evi-dent) by the investigators were confirmed to be negativeby the histologic results. The positive predictive value was39.2% (20/51). Thus, only 39.2% of the patients assessed aspositive by the investigators (target tumor clinically evi-dent) were confirmed to be positive by the histologicresults. The sensitivity of the investigator assessment ofpositive sites was 100% (20/20). This indicated that theinvestigators correctly identified as positive all of the pa-tients who were confirmed to be positive by the histology.In some cases, clinical clearance was not determined.

Three patients were not assessed by the investigatorfor tumor clearance. In 4 patients, the investigatorcould not determine whether tumor was present; in all4 of these cases, there was no histologic evidence ofresidual tumor.

Adverse eventsAdverse events occurred in all treatment groups,

with 118 (92.2%) of 128 patients reporting at leastone adverse event. Application site reactions, in-cluding itching, pain, and tenderness at the targettumor site, were the most frequently reported ad-verse events (Fig 7). Thirty-two severe adverseevents (ie, events that substantially interfered withdaily activities) were reported by 25 patients; 6 wereconsidered to be probably or possibly related to

Fig 1. Proportion of patients who were complete responders based on histologic examinationof posttreatment excision tissue.

Fig 2. Superficial BCC (target tumor) before treatment. Fig 3. Target tumor area after 4 weeks of daily treatmentwith imiquimod 5% cream.


SEPTEMBER 2002

study drug (all 6 were application site reactions inthe once daily treatment group). Six serious adverseevents (Shy Drager syndrome [postural hypoten-sion], fever and lobar pneumonia, Staphylococcusinfection of the right elbow [unrelated to treatmentsite on left side of chest], transient ischemic attack,and a prostate biopsy) were reported by 5 patientsand resulted in hospitalizations. These hospitaliza-tions were related to intercurrent illnesses or eventsand were not considered to be related to study drug;all affected patients recovered and completed thestudy. Four patients discontinued dosing with studydrug early because of treatment-related adverseevents (3 because of pain at the treatment site, 1 ofwhich was severe, and 1 because of bleeding at thetreatment site).

Local skin reactionsLocal skin reactions, identified by type as erythema,

scabbing, erosion, excoriation/flaking, induration,edema, ulceration, and vesicles occurred in all treat-ment groups (Fig 8). In the twice-daily group, severeerythema and severe scabbing were each recorded byinvestigators for 3 (30%) of the 10 patients, and severeerosion and severe excoriation/flaking were each re-corded for 1 (10%) of the 10 patients. In the othertreatment groups, erythema was noted most oftenamong severe reactions. Of the 13 patients discontin-ued from treatment because of local skin reactions(listed as the primary reason), 6 patients had local skinreactions considered severe (Fig 9).

DISCUSSIONThis study demonstrated that 87.1% and 80.8% of

patients who treated a single, primary, nonaggres-

sive sBCC with imiquimod 5% cream once daily or 5times a week for 12 weeks were histologically free oftumor 6 weeks after treatment. Of the patients whotreated similar tumors 3 times a week for 12 weeks,51.7% were histologically free of tumor. An accept-able safety profile was seen in 3 of the 4 imiquimoddosing regimens. Only the most frequent dosingregimen, twice daily for 12 weeks, presented a safetyprofile that was judged not acceptable because ofsevere local skin reactions at the treatment site. Aclearance rate of 18.8% was also noted in thevehicle treatment group and was remarkably consis-tent with the tumor cure rate previously reported forthe placebo-treated group of a controlled study us-ing intralesional interferon for the treatment of smallBCC tumors that were similarly diagnosed.13 As ex-pected, the vehicle response rate in this study, inwhich the size of the diagnostic biopsy was limited,was lower than a recently reported histologic curerate of 33% observed approximately 1 to 6 monthsafter standard diagnostic punch or shave biopsies of42 small, primary BCC tumors.16

Superficial BCC was confirmed with a small, pre-

Fig 4. Target tumor area 6 weeks after imiquimod treat-ment, before excision; no clinical evidence of BCC (con-firmed clear histologically).

Fig 5. Superficial BCC evident in target tumor biopsybefore treatment.

Fig 6. No evidence of BCC in section of target tumor areaexcision specimen 6 weeks after treatment.


VOLUME 47, NUMBER 3

study biopsy (�25% of the clinically evident tumor)before treatment, in accordance with standard derma-tologic practice. Only those tumors with superficial BCC(with no nodular or aggressive features) were intendedfor treatment in this study. Because of the small size ofthe specimen, the prestudy biopsy may not have beenrepresentative of the histology of the entire tumor, and12 patients with mixed superficial and nodular histologywere included in the study. Although the number ofpatients with mixed histology enrolled into each treat-ment group was not analyzed separately, 9 of the 12patients (including all 6 of those treated once daily or 5times a week) were complete responders to treatment,indicating that the response of superficial tumors withsome nodular histologic components may be compara-ble to the response of tumors with purely superficialhistology on biopsy.

A posttreatment excision of the entire target tumorarea was performed, and the excised tissue samplewas examined for any histologic evidence of residualtumor. The step-sectioning procedure was much morethorough than standard sectioning, and it is unlikelythat any residual tumor was missed. The inclusion of aposttreatment excision was considered appropriate forthis stage of clinical development because the efficacyof imiquimod for BCC has not been fully defined. Theposttreatment excision of the tumor area allowed pa-tients to have their tumors surgically removed if thestudy drug was not effective, as well as providingdefinitive evidence of tumor clearance. The posttreat-ment excision and its thorough histologic evaluationserved as the endpoint for the treatment effect of imi-quimod cream, rather than the clinical assessment oftumor clearance or persistence.

Table III. Summary statistics of the investigator assessment of the target tumor site

VariableAll imiquimod

groups combined

Imiquimod dose group

3 times a week 5 times a week Once daily Twice daily

Negative predictive value 100% (38/38) 100% (9/9) 100% (8/8) 100% (16/16) 100% (5/5)Positive predictive value 39.2% (20/51) 70% (14/20) 20% (3/15) 25% (3/12) 0% (0/4)Prevalence 22.5% (20/89) 48.3% (14/29) 13% (3/23) 10.7% (3/28) 0% (0/9)Sensitivity 100% (20/20) 100% (14/14) 100% (3/3) 100% (3/3) — (0/0)Specificity 55.1% (38/69) 60% (9/15) 40% (8/20) 64% (16/25) 55.6% (5/9)

Fig 7. A, Patients reporting application site reactions. In the once daily group, 5 patients (16%)reported severe application site reactions. B, Most frequently reported application site reac-tions.


SEPTEMBER 2002

The clinical assessment for persistent tumor done byinvestigators 6 weeks after treatment, however, wasilluminating. When the investigators recorded no clini-cal evidence of BCC 6 weeks after treatment, histologyconfirmed the clinical impressions in all 38 cases (100%negative predictive value). Of 51 cases in which theinvestigators had recorded clinical evidence of BCC atthe treated target tumor site, only 20 had BCC confirmedhistologically (39% positive predictive value). These 20represented all of the histologically positive cases in thestudy, indicating that the investigators were successfullyable to identify residual BCC by clinical evaluation in allof the treatment failures. In 31 cases, the investigatorsrecorded evidence of BCC, which was not supported bythe histology. This indicated that investigators had somedifficulty determining that the tumor was gone in apercentage of cases that were actually clear. It is as-sumed that this difficulty is related to the persistence ofsome low-grade erythema in a percentage of cases at 6weeks after treatment when the assessments were done.

Consistent with past studies14,17 using imiquimod 5%cream for the treatment of BCC, the most commonlyreported adverse event was itching at the target site.Pain and tenderness at the target site were the next mostfrequently reported symptoms.

Despite the fact that nearly 20% of patients discon-tinued their treatment before the end of the planned12-week treatment period, several of these patients inthe twice-daily, once-daily, and 5-times-a-week treat-ment groups were completely clear of BCC on histo-logic examination. In another study evaluating 33 pa-tients, dosing once daily with imiquimod for 6 weeks

(rather than the 12-week treatment in this study) thecomplete response rate was 87.9%.18 This suggeststhat additional clinical studies are needed to deter-mine the appropriate duration of treatment for thisnovel and effective therapy.

Mohs micrographic surgery offers the best curerates for the treatment of BCC. The recurrence rateswith surgical excision for primary BCCs vary basedupon the characteristics and location of the tumortreated and the size of the surgical margin, but ingeneral range from 3.2% to 8%, so that the cure ratesare 92% to 96.8%.19 The 5-year cure rates are similarfor curettage and cautery.20 The complete response

Fig 8. Investigator-assessed local skin reactions. Reactions were not seen in all dosegroups.

Fig 9. Severe local reactions after 4 weeks of twice-dailyimiquimod treatment.


VOLUME 47, NUMBER 3

rate of 87.1% seen in the once-daily imiquimodgroup is lower than the response rate seen in thesecommonly used modalities for primary BCC. How-ever, with the incidence of BCC increasing, imi-quimod may still offer an alternative treatment thatavoids surgical or ablative treatment.19

Recurrent BCC after incomplete resection and thepresence of residual BCC in excisional margins con-tinue to present a treatment challenge.21 Although cu-rettage before excision decreased margin positivity af-ter excision by 24%, this procedure still results in afailure rate with positive histologic margin in 13% ofsuch excisions.22 The potential use of imiquimod as anadjuvant treatment to surgery may warrant consider-ation. It may be beneficial to determine whether imi-quimod could reduce the size of larger tumors beforesurgery, and to see whether the immune system couldinduce a specific TH1 immune response that wouldperhaps decrease the risk for recurrence of high riskand recurrent tumors. Clearly, with larger studies, imi-quimod may be proven as a viable alternative to sur-gery for low-risk sBCC tumors.

In conclusion, imiquimod 5% cream was effectivein the treatment of sBCC in all dosing regimens usedin this study. The once-daily or 5-times-a-week dos-age groups had the highest efficacy results withacceptable safety profiles. This novel therapy offersa potential nonsurgical option for a disease tradi-tionally treated by surgery and may be important inour management of an increasingly common cuta-neous malignancy.

The following are principal investigators (study centersare given in parentheses): John K. Geisse, MD, coordinat-ing investigator (Solano Dermatology Associates, Vallejo,Calif); Libby Edwards, MD (Carolinas Medical Center,Charlotte, NC); Rokea A. el-Azhary, MD (Mayo Clinic,Rochester, Minn); Kenneth Gross, MD (Skin Surgery Med-ical Group, Inc, San Diego, Calif); Hubert T. Greenway,MD (Scripps Clinic, La Jolla, Calif); Norman Levine, MD(University of Arizona, Tucson); Ida Orengo, MD (BaylorCollege of Medicine, Houston, Tex); Phoebe Rich, MD(Northwest Cutaneous Research Specialists, Portland,Ore); June K. Robinson, MD (Loyola University MedicalCenter, Maywood, Ill); Amit Pandya, MD (UT SoutwesternMedical Center, Dallas, Tex); Elizabeth Shurnas, MD, andPaul Brownstone, MD (Alpine Clinical Research Center,Boulder, Colo); James Swinehart, MD (Colorado MedicalResearch Center, Denver); and David Tashijian, MD (Cen-tral California Medical Research, Fresno).

We thank Lawrence Gibson, MD (Mayo Dermatopa-thology, Rochester, Minn) and Loren Golitz, MD (Der-matopathology Services, L.L.C., Denver, Colo) for his-topathologic analyses; Aimee Couture for studymanagement; Terry Fox for statistical analyses support;Jody Hubred for data management; and Altha Edgren formanuscript preparation.

REFERENCES

1. Greenlee RT, Murray T, Bolden S, Wingo PA. Cancer statistics,2000. CA Cancer J Clin 2000;50:7-33.

2. Miller DL, Weinstock MA. Nonmelanoma skin cancer in theUnited States: incidence. J Am Acad Dermatol 1994;30:774-8.

3. Puizina-Ivi N, Matokovi B, Gluni I, Maslovara S, Vela-Ljubi J. His-topathologic variants of basal cell carcinoma correlation withsex, age and localization. J Med Syst 1999;23:389-400.

4. Bastiaens MT, Hoefinagel JJ, Bruijn JA, Westendorp RGJ, VermeerBJ, Bavinck JNB. Differences in age, site distribution, and sex be-tween nodular and superficial basal cell carcinomas indicate differ-ent types of tumors. J Invest Dermatol 1998;110:880-4.

5. Jaramillo-Ayerbe F. Cryosurgery in difficult to treat basal cell car-cinoma. Int J Dermatol 2000;39:223-9.

6. McCormack CJ, Kelly JW, Dorevitch AP. Differences in age andbody site distribution of the histological subtypes of basal cellcarcinoma: a possible indication of differing causes. Arch Der-matol 1997;133:593-6.

7. Silverman MK, Kopf AW, Grin CM, Bart RS, Levenstein MJ. Recur-rence rates of treated basal cell carcinomas. Part 1: overview;Part 2: curettage-electrodesiccation. J Dermatol Surg Oncol1991;17:713-8, 720-6.

8. Silverman MK, Kopf AW, Grin CM, Bart RS, Levenstein MJ. Recur-rence rates of treated basal cell carcinomas. Part 3: surgical ex-cision. J Dermatol Surg Oncol 1992;18:471-6.

9. Robinson JK. Risk of developing another basal cell carcinoma: a5-year prospective study. Cancer 1987;60:118-20.

10. Slade HB, Owens ML, Tomai MA, Miller RL. Imiquimod 5% cream(Aldara). Exp Opin Invest Drugs 1998;7:437-50.

11. Edwards L, Whiting D, Rogers D, Luck K, Smiles KA. The effect ofintralesional interferon gamma on basal cell carcinomas. J AmAcad Dermatol 1990;22:496-500.

12. Edwards L, Tucker S, Perednia D, Smiles KA, Taylor EL, Tanner DJ,et al. The effect of an intralesional sustained-release formulationof interferon alfa-2b on basal cell carcinomas. Arch Dermatol1990;126:1029-32.

13. Cornell RC, Greenway HT, Tucker SB, Edwards L, Ashworth S,Vance JC, et al. Intralesional interferon therapy for basal cell car-cinoma. J Am Acad Dermatol 1990;23:694-700.

14. Beutner KR, Geisse JK, Helman D, Fox TL, Ginkel A, Owens ML.Therapeutic response of basal cell carcinoma to the immuneresponse modifier imiquimod 5% cream. J Am Acad Dermatol1999;416:1002-7.

15. Fitzpatrick TB. The validity and practicality of sun-reactive skintypes I through VI. Arch Dermatol 1988;124:869-71.

16. Holmkvist KA, Rogers GS, Dahl PR. Incidence of residual basalcell carcinoma in patients who appear tumor free after biopsy.J Am Acad Dermatol 1999;41:600-5.

17. Kagy MK, Amonette R. The use of imiquimod 5% cream for thetreatment of a superficial basal cell nevus syndrome patient.Dermatol Surg 2000;26:577-9.

18. Marks R, Gebauer K, Shumack S, Amies M, Bryden J, Fox TL, et al.Imiquimod 5% cream in the treatment of superficial basal cellcarcinoma: results of a multicenter 6-week dose-response trial.J Am Acad Dermatol 2001;44:807-13.

19. Telfer NR, Colver GB, Bowers PW. Guidelines for the manage-ment of basal cell carcinoma. Br J Dermatol 1999;141:415-23.

20. Thissen MR, Neumann MH, Schouten LJ. A systematic review oftreatment modalities for primary basal cell carcinomas. ArchDermatol 1999;135:1177-83.

21. Robinson JK, Fisher SG. Recurrent basal cell carcinoma after in-complete resection. Arch Dermatol 2000;136:1318-24.

22. Chiller K, Passaro D, McCalmont T, Vin-Christian K. Efficacy ofcurettage before excision in clearing surgical margins of non-melanoma skin cancer. Arch Dermatol 2000;136:1327-32.


SEPTEMBER 2002

Imiquimod 5% cream for the treatment of superﬁcial basal ... 5%...Imiquimod 5% cream for the treatment of superﬁcial basal cell carcinoma: A double-blind, randomized, vehicle-controlled

Documents