FEASIBILITY OF 5-FLUOROURACIL AND IMIQUIMOD FOR THE TOPICAL TREATMENT OF CERVICAL INTRAEPITHELIAL NEOPLASIAS (CIN) 2/3

Abstract

OBJECTIVES:

Determine the feasibility (as measured by tolerability and safety) and efficacy of topical 5-fluorouracil (5-FU) and imiquimod for treatment of cervical intraepithelial neoplasia (CIN) 2/3.

METHODS:

This pilot prospective study was conducted in women aged 18–45 with p16+ CIN 2/3. Participants underwent an 8-week alternating regimen of self-applied 5% 5-FU on weeks 1/3/5/7 and physician-applied imiquimod on weeks 2/4/6/8. Adverse events (AEs) were collected by symptom diary and clinical exam. Feasibility was measured by tolerability and safety (AE’s) of the study intervention. Tolerability was assessed as the number of participants able to apply >=50% of the treatment doses. The safety outcome was calculated as the number of participants who experienced ‘specified AEs’ defined as possibly, probably, or definitely related grade 2 or worse adverse event or grade 1 genital AEs (blisters, ulcerations, or pustules) lasting >5 days. Efficacy of the intervention was determined by histology and high-risk HPV (hrHPV) testing were collected after treatment.

RESULTS:

Median age of the 13 participants was 27 ± 2.9 years. Eleven (84.61%) participants applied >=50% of the treatment. All participants reported grade 1; 6 (46.15%) grade 2; and 0 had grade 3/4 AEs. Three (23.08%) participants had specified AEs. Histologic regression to normal or CIN 1 among those completing >= 50% treatment doses was observed in 10 (90.91%) participants, and 7 (63.63%) tested negative for hr-HPV at study end.

CONCLUSIONS:

Topical treatment for CIN 2/3 with 5-FU/imiquimod is feasible, with preliminary evidence of efficacy. Topical therapies need further investigation as adjuncts or alternatives to surgical therapy for CIN 2/3.

Clinical Trial Number: NCT03196180

Clinicaltrials.gov https://www.clinicaltrials.gov/ct2/show/NCT03196180?term=NCT03196180&draw=2&rank=1

INTRODUCTION:

Cervical cancer is the 4^th leading cancer in women worldwide and the 8^th most common cancer in women in the United States [1, 2]. The majority of the global burden of invasive cervical cancer (ICC) and its precursor lesions occurs in low- and middle-income countries (LMICs).¹⁴ In sub-Saharan Africa, where the vast majority of cervical disease occurs, ICC is the leading cause of cancer death among women [1, 3]. Cervical cancer screening and treatment prevention programs consist of screening with cytology and/or high risk human papillomavirus (hr-HPV) testing followed by colposcopy, which detects high grade precursor lesions, or cervical intraepithelial neoplasia (CIN) 2/3 (‘CIN 2/3’). These lesions are then treated with surgical excision [loop excision electrosurgical procedure (LEEP), cold knife cone (CKC)] or ablation (laser therapy or cryosurgery) [4]. Though excision and ablation are generally effective approaches, there are associated risks with these treatment options including pain, bleeding, and increased risk of preterm birth [5, 6].

The disparity in ICC incidence and mortality between LMICs and industrialized countries can largely be attributed to differences in healthcare access, particularly for cancer screening and treatment [7]. Surgical treatment can be difficult to access or unavailable. Women living with HIV (WLWH) experience high rates of CIN 2/3 and cervical cancer, particularly in LMICs due to inadequate access [8–10] as well as increased risk of CIN recurrence after standard surgical treatment [11, 12].

There are no current recommendations for non-surgical management of CIN 2/3. Two CIN 2/3 treatments being evaluated are 5-fluorouracil (5-FU), an antiproliferative agent, and imiquimod, an immune modulator [13–18]. Although they have been used individually for the treatment of genital warts, vaginal/vulvar lesions, and have been explored individually for the treatment of CIN 2/3 [13–15], neither their efficacy nor their safety in combination is known.

The primary aim of this study was to determine the feasibility of an alternating regimen of intravaginal 5-FU and imiquimod for the treatment of CIN 2/3. A secondary aim was to assess efficacy. The use of medical options such as 5-FU and imiquimod could potentially help address service delivery constraints in LMIC settings by allowing immediate medical treatment when surgical treatment is difficult to access or unavailable; by repurposing widely available, low-cost, generic drugs; and by mitigating the cost and burden on the patient and health system.

MATERIALS AND METHODS:

We conducted a phase 1 study of the feasibility (as measured by tolerability and safety) of topical 5-FU and imiquimod in women with CIN 2/3 from 09/30/2019 through 11/04/2020 in a single healthcare system with a passive post-intervention surveillance period to 12/31/21. The University of North Carolina (UNC) Institutional Review Board and the National Cancer Institute Division of Cancer Prevention (NCI, DCP) approved this study. Signed patient consents were obtained from all participants.

We enrolled cis-gender English- or Spanish-speaking women 18–45 years old who were diagnosed with p16 positive CIN2 or CIN3 on histology. We excluded women who had 1) been previously treated with 5-FU or imiquimod, 2) had concurrent vaginal, vulvar, or anal lesions, or 3) had a prior hysterectomy. We also excluded women with any invasive malignancy in the last five years and women with certain pathologic diagnosis (including atypical glandular cells, cervical carcinoma on Pap smear or biopsy, or CIN 3 on more than two cervical quadrants). Additionally, we excluded women who were on anticoagulants or used investigational agents six months prior to enrollment, had known dihydropyrimidine dehydrogenase (DPD) enzyme deficiency, or reported a history of allergic reaction attributed to compounds of similar chemical or biologic composition as the study medications. Pregnant women, those planning on pregnancy, those unwilling to use condoms plus another method of contraception, or women breastfeeding in the next six months were also excluded.

Medication Regimen

We designed the study with planned application of 5-FU and imiquimod once weekly on alternating weeks over two 8-week phases of the study. Participants self-applied 2 grams of 5-FU cream (5% 5-fluorouracil, Effudex, Bausch Health, Laval, Canada) at home using an applicator (weeks 1, 3, 5, and 7 corresponding with dose numbers 1, 3, 5, 7) for 4 applications. Participants also presented for physician-administered imiquimod (Aldara, Bausch Health, Laval, Canada) in weeks 2, 4, 6, and 8, corresponding with dose numbers 2, 4, 6, 8. Study physicians (L.R. and N.D.) directly applied Imiquimod (12.5 mg of 5% imiquimod in 250 mg cream) to the cervix on the weeks following the 5-FU dose so that each dose of either medication was at least 7 days apart.

The study’s second phase was optional if a participant wanted to continue topical treatment for an additional 8 weeks (total intervention of 16 weeks). We maintained the same medication regimen with 5-FU self-application on weeks 9, 11, 13, and 15 and physician-applied imiquimod on weeks 10, 12, 14, and 16. In both phases of the study, a participant reporting menses that coincided with an application date could delay application until the end of a menstrual cycle. There was always a minimum 7-day interval between application days. We required participants to avoid sexual activity for 48 hours after applying the study medication.

Study Visits

The study coordinators pre-screened women with CIN 2/3 by phone and scheduled a screening visit to obtain informed consent. The participants were then scheduled for a baseline visit during which study physicians (L.R. and N.D.) performed a colposcopic exam with digital imaging of the cervix. Study coordinators educated participants on the use of the 5-FU cream and provided them 5-FU cream, vaginal applicators, condoms, pregnancy test kits, and written instructions on self-applying the 5-FU cream.

We instructed the women to use the 5-FU cream on weeks 1, 3, 5, and 7 at home. We reviewed their symptom diary on the subsequent study visits (weeks 2, 4, 6, 8) and performed a colposcopic examination of the cervix with digital imaging followed by imiquimod application. We defined and assessed safety using the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTCAE) version 4.0 for non-genital events and Female Genital Grading Table (FGGT) as appropriate. Study coordinators weighed the unused applicators and the tube of 5-FU cream to confirm decreasing amounts of residual cream at every study visit. After completing the first 8-week phase of the study, we offered the participants the option to continue for an additional 8 weeks. Those who opted to enter this extension phase had 4 additional self-applications of 5-FU (weeks 9, 11, 13,15) and physician-applied imiquimod (weeks 10, 12, 14,16) by study physicians (L.R. and N.D.).

Participants returned for an end-of-study visit four to six weeks after the last topical application. The study physician (L.R.) performed a colposcopic exam of the cervix. Pap smear (ThinPrep, Hologic, Marlborough, MA, USA), hr-HPV testing (Thin Prep, Hologic, Massachusetts, USA), and at least two colposcopy-directed biopsies were collected. A single board-certified GYN pathologist (S.O.) evaluated cytology and histology. We used the current clinical guidelines to offer clinical management (LEEP for persistent CIN 2/3; 6 months follow-up for all other results) [4].

Statistical Plan

In this pilot study, our primary aim was to determine the feasibility of dual intravaginal therapy for 5-FU and imiquimod as measured by tolerability and safety. Tolerability was defined as the ability to apply ≥ 50% of the treatment (i.e. at least 4 weeks or 2 doses of 5-FU and 2 doses or imiquimod). We defined our safety outcome based on the number of participants experiencing specified adverse events. Specified adverse events (AEs) were defined as Grade 2 AEs or Grade 1 genital lesions (blisters, ulcerations, or pustules) that were possibly, probably, or definitely related and lasted greater than 5 days. Of note, in the original protocol published on Clinicaltrials.gov, the term “dose-limiting toxicity” is listed incorrectly instead of “specified adverse outcome.” The study was designed to evaluate the feasibility of further investigation of this dual therapy strategy, with an apriori set decision of moving forward if the specified AEs were observed in less than 33% of participants and tolerability was achieved in more than 66% of participants. With a sample size of at least 12 participants, if the projected specified AEs was 30%, the study would be able to detect a specified AEs in 33% or more of participants with a probability of 50.7%. Similarly, if the projected tolerability rate was 50%, the study with 12 participants would have been able to observe tolerability in 66% or less participants with a probability of 80.6%. Baseline participant characteristics, AEs, tolerability are reported using descriptive statistics. We also reported efficacy, defined as histologic regression or clearance of hr-HPV.

RESULTS:

Fourteen women agreed to participate in this study. One participant never started treatment, and thus 13 participants were utilized for analysis. The median age was 27 years [IQR ± 2.9 years]; 9 (69.23%) were less than 30 years old. Baseline participant characteristics are outlined in Table 1. The length of therapy ranged from 11 days to 144 days. Of the 13 participants enrolled, 11 completed the first 8-week phase of the study, and of those, 6 opted to continue for an additional 8 weeks. Two out of 13 participants withdrew; one after the first dose (moved out of state) and another after three doses (had AE and concerns regarding COVID-19 pandemic restrictions). Eleven (84.61%) participants were able to use at least 50% of the treatment.

Table 1.

Baseline characteristics in a cohort of women with CIN 2,3 in a feasibility study of alternating 5FU and imiquimod for topical treatment (n=13).

	Study Cohort N (%)
Age
Median (range)	27 (22 – 35)
Race/Ethnicity
White/Non-Hispanic	7 (53.85%)
Black	3 (23.08%)
Hispanic	2 (15.38%)
Asian	1 (7.69%)
BMI (kg/m2)
Underweight (BMI < 18)	0 (0.0%)
Normal Weight (BMI 18.0 – <25.0)	6 (46.15%)
Overweight (BMI 25.0 - < 30.0)	1 (7.69%)
Obese (BMI >= 30.0)	6 (46.15%)
Tobacco Use
No prior Tobacco Use	11 (84.62%)
Former Tobacco Use	1 (7.69%)
Current Tobacco Use	1 (7.69%)
Alcohol Use in the past 12 months	12 (92.31%)

While all 13 (100%) participants reported grade 1 AEs, and 6 (46.15%) noted grade 2 AEs, none (0%) of the participants experienced grade 3 or 4 AEs. The most common AEs were vaginal/lower abdominal/pelvic pain or discomfort [10 (76.92%)] and vaginal discharge [7 (53.84%)]. Five participants reported systemic symptoms such as fever (1), myalgia (1), and headache (4), not mutually exclusive (Table 3). Three (23.08%) participants experienced the specified AEs which defined the safety outcome – two (14%) participants with a grade 2 vaginal lesion (defined as size < 1 cm), and one with grade 2 vaginal discharge. All specified AEs resolved spontaneously.

Table 3.

Summary of study-related adverse events (AE) (n=13 participants)

	Grade 1a N (%)	Grade 2a N (%)a
Bacterial Vaginosis	0 (0%)	1 (7.69%)
Candida	0 (0%)	1 (7.69%)
Cervical friability	1 (7.69%)	0 (0%)
Cervical/vaginal erythema	1 (7.69%)	0 (0%)
Dyspareunia	1 (7.69%)	0 (0%)
Dysuria	1 (7.69%)	0 (0%)
Fever	1 (7.69%)	0 (0%)
Headache	3 (23.08%)	1 (7.69%)
Myalgia	0 (0%)	1 (7.69%)
Pain	8 (61.54%)	2 (15.38%)
Pruritus	4 (30.77%)	1 (7.69%)
Unexplained bleeding	4 (30.77%)	0 (0%)
Upper respiratory infection	5 (38.46%)	0 (0%)
Vaginal abrasion/lesion	2 (15.38%)	2 (15.38%)
Vaginal discharge	5 (38.46%)	2 (15.38%)

^ahighest grade if a participant had AE more than once

We also assessed response to therapy by either histologic regression or clearance of hr-HPV (negative hr-HPV result as end of study visit). Twelve of 13 (92.31%) participants completed the end-of-study visit. Ten of these 12 (83.33%) participants demonstrated histologic regression (8 with normal histology and 2 with CIN 1). Seven (58.33%) participants had clearance of their hr-HPV. Of the 11 women who took at least 4 weeks of medication (i.e. tolerated the medication as per our definition of tolerability), 90.91% had histologic regression, and 63.64% had hr-HPV clearance (Table 2). Normal/overweight (i.e.,BMI < 30) participants had higher hr-HPV clearance rates (83.33%) than obese (i.e., BMI >= 30) participants (16.66%). There was no association between tobacco use and hr-HPV or CIN 2/3 at the end of treatment, p = 1.00.

Table 2:

Participants with adverse events (AE), tolerability, and treatment response (n=13)

Outcome	N (%)	95% CIf
Grade 1 AE	13 (100%)	NA
Grade 2 AE	6 (46.15%)	(19.22%, 74.87%)
Specified adverse events a	3 (23.08%)	(5.04%, 53.81%)
Tolerability b	11 (84.62%)	(54.55%, 98.08%)
Response c
Histologic regression	10 (83.33%)	(51.59%, 97.91%)
Histologic regression amongst participants who tolerated treatmentd	10 (90.91%)	(58.72%, 99.77%)
High-risk HPV clearancee	7 (58.33%)	(27.67%, 84.83%)
High-risk HPV clearancee amongst participants who tolerated treatmentd	7 (63.64%)	(30.79%, 89.07%)

^anumber of participants with AEs defined as Grade 2 or greater toxicity or Grade 1 genital lesion (blisters, ulcerations, or pustules) that was possibly, probably, or definitely related to treatment and lasted for more than 5 days; all had Grade 2

^bnumber participants able to apply ≥ 50% of the treatment (i.e. at least 4 weeks)

^cResponse was measurable in 12 out of 13 participants who completed end of study visit

^d11 participants who were able to apply ≥50% treatment (i.e at least 4 weeks)

^eNegative HPV DNA testing at end of study visit

^fClopper-Pearson (“exact”) 95% CI

DISCUSSION:

This study sought to evaluate the feasibility of topical therapy for the treatment of CIN 2/3 defined by tolerability and safety. The regimen of self-applied topical 5-FU and provider-applied topical imiquimod was able to be successfully administered and tolerated amongst most women despite all women reporting AEs and approximately 25% reporting Grade 2 specified-AEs. This study suggests that women interested in alternative therapies to surgical excision for the treatment of CIN 2/3 would tolerate a combined 5-FU/imiquimod regimen.

The side effect profile of the combined 5-FU/imiquimod regimen was similar to prior single-agent side effect profile of topical 5-FU or imiquimod. Two prior randomized trials of self-applied topical 5-FU monotherapy given once every 2 weeks reported that 33 to 48% of participants experienced Grade 1 or 2 AEs [16, 17]. A randomized trial of escalating doses of imiquimod delivered by vaginal suppositories reported 92% and 97% of participants with Grade 1 local side effects and systemic side effects (flu-like symptoms and fatigue), respectively [13]. A recent trial of provider-applied weekly imiquimod reported 78% of participants with Grade 1–2 side effects [19].

In our study, 90.91% of participants demonstrated histologic regression, and 63.64% had negative hr-HPV testing if they used more than 4 weeks (2 doses of 5-FU and 2 doses of imiquimod) of treatment. Previous research on 5-FU as monotherapy for CIN 2 reported 93% histologic regression and 42.9% with hr-HPV-clearance [17]. Primary treatment of CIN 2–3 with escalating doses of vaginal imiquimod demonstrated 73% histologic regression and 60% hr-HPV-clearance [13]. Weekly provider-applied imiquimod to CIN 2–3 lesions reported 61% histologic regression [19]. While we cannot directly compare these studies to our feasibility study due to differing treatment regimens and statistical power, our study demonstrates a similar efficacy signal as the above regimens.

The strengths of this study are its innovative approach to the treatment of cervical dysplasia. To date, there are no approved medical therapies, and this represents a substantial unmet medical need for women who may not be appropriate candidates for treatment by excisional or ablation therapy. This may also benefit LMIC’s with limited resources or substantial patient barriers to excisional therapies. Another strength is that participants successfully self-applied the topical 5-FU, which allowed for convenient home administration and averted multiple pelvic examinations. Medical treatment and self-administered topical therapy may be useful as primary treatment in resource-limited or rural settings, or as pre- or post-surgical adjuvant therapy to decrease lesion size, treat positive margins, and prevent recurrence. Of note, data from this pilot study are being used to inform current topical therapy intervention trials in LMIC settings (Clinical trial.gov Identifier NCT05413811, NCT05362955).

Limitations of the study include a small sample size and single study site which could lead to selection bias. Additionally, given the combination regimen, we cannot attribute AEs to a particular drug, though imiquimod is known to confer more systemic AEs compared to topical 5-FU. Lastly, we were unable to adequately assess the contribution of tobacco use to HPV clearance rates in the setting of medical treatment with our small cohort. Our data also signals that obesity may impact HPV clearance, but an associated pathophysiology is elusive given that it was topical therapy. Patient factors will need to be further assessed in future prospective trials.

In summary, this study demonstrated that participants tolerated dual therapy of imiquimod and 5-FU despite common Grade 1 or 2 AEs. Our hr-HPV clearance and histologic regression rates are a positive signal to consider a larger clinical trial to quantify regression and clearance rate. Though it is unclear if combination therapy is superior to treatment with imiquimod or 5-FU alone, medical treatment for CIN 2/3 is an endeavor worthy of further exploration and would add another intervention to our cervical cancer prevention toolkit that is immediately accessible across the world.

SYNOPSIS:

Topical application of alternate-weekly dosing of 5-flurouracil/imiquimod is feasible, with preliminary evidence of efficacy, among women seeking alternatives to surgical excision for treatment of cervical precancer.

ABBREVIATIONS AND ACRONYMS:

5-FU

5-fluorouracil

AE

Adverse Event

CIN

cervical intraepithelial neoplasia

CKC

Cold knife cone

CTCAE

Common Terminology Criteria for Adverse Events

DPD

dihydropyrimidine dehydrogenase

FGGT

Female Genital Grading Table

HR

High risk

HPV

Human papillomavirus

LEEP

Loop electrosurgical excision procedure

NCI

National Cancer Institute