Vulvovaginal Lichen Planus Treatment Outcomes: A Systematic Review of Novel and Established Therapies

Protocol and Registration

This systematic review was registered with PROSPERO (CRD42025101229) prior to study commencement. The protocol is publicly available at https://www.crd.york.ac.uk/prospero/. The risk of bias (RoB) assessment tools (Newcastle-Ottawa Scale for case series and Cochrane RoB tool for randomized controlled trials [RCTs]) were specified during the protocol refinement phase. No other amendments were made to the registered protocol.

Eligibility Criteria

We structured eligibility criteria using the PICO framework.

The population included adults (≥18 years) with clinically or histologically confirmed VVLP (with or without gingival involvement), encompassing all disease variants. Interventions included pharmacological treatments (all routes), procedural interventions (laser, photodynamic therapy, surgery), and supportive therapies. Studies with or without comparison groups were eligible.

Primary outcomes were complete resolution, partial resolution, no response, recurrence rates, and adverse events. Secondary outcomes included time to response, quality of life, and treatment discontinuations. We included RCTs, non-RCTs, cohort studies, case-control studies, case series (≥3 patients), and case reports, excluding reviews, editorials, and abstracts. No date restrictions were applied (inception to February 28, 2025). No language restrictions were applied to the search strategy; however, 7 studies identified during screening were published in languages for which translation resources were unavailable, and these were excluded at full-text review. This is acknowledged as a limitation. For recurrence outcomes, a minimum 1-month follow-up was required.

Outcome Definitions

Given the absence of standardized outcome measures, we used operational definitions based on the primary study reporting.

Complete resolution was defined as complete clearance of lesions and symptoms (pain, burning, discharge, dyspareunia) on clinical examination or patient report, or return to baseline when scoring systems were used. Partial resolution indicated improvement without complete clearance. No response indicated absent improvement or disease progression. Recurrence was the re-emergence of lesions or symptoms after resolution, with the time calculated from resolution to the detection of recurrence. Adverse events included any unfavorable occurrence regardless of causality, categorized by severity when provided. We extracted data from the longest available follow-up (range 2-104.4 months) and reported follow-up duration individually for each study.

Most included studies did not clearly differentiate between vaginal-only, vulvar-only, or combined vulvovaginal disease. Disease location was typically reported as ‘vulvovaginal lichen planus’ without specifying the anatomical distribution of lesions. This limitation affects the interpretation of site-specific treatment approaches.

Search Strategy

We searched MEDLINE, Embase, CINAHL, and Scopus (inception to February 28, 2025) using terms for VVLP and treatments. No language restrictions were applied to the database searches. Seven studies published in languages other than English were excluded at the full-text review stage due to unavailable translation resources. Full strategies are in Supplemental Table 1. Reference lists of included studies were screened for additional relevant publications.

Gray literature searching was not systematically conducted. Pilot searches suggested VVLP studies are predominantly published in peer-reviewed journals, and comprehensive gray literature searching is resource-intensive. However, we performed forward and backward citation searching.

A medical librarian (L.D.) developed and executed the search strategy, peer-reviewed using the PRESS checklist.

Study Selection and Data Extraction

Titles, abstracts, and full texts were independently screened and extracted by 2 reviewers, with disagreements resolved with a third reviewer. Extracted data included the following: study design, patient demographics, intervention details (dose, duration, delivery), and outcomes such as complete resolution (CR), partial resolution (PR), no response (NOR), adverse events, treatment discontinuations, and recurrence.

Software and Tools

Screening and data management were conducted using Covidence systematic review software. Data were independently extracted by 2 human reviewers, with discrepancies resolved through discussion. Grammarly was used as a writing assistance tool for grammar and readability during manuscript preparation; all suggestions were reviewed and approved by the authors. No artificial intelligence tools were used for any analytical component of this review, including screening, data extraction, or data interpretation.

Full-Text Articles

During full-text retrieval, 35 articles could not be located despite comprehensive efforts (author requests, interlibrary loan, librarian consultation). These were primarily older publications (pre-2000) or journals with limited access. All unretrieved articles and excluded studies with reasons for exclusion are documented in Supplemental Table 6.

RoB Assessment

RoB was independently assessed by 2 reviewers using the Cochrane RoB tool for RCTs and the Newcastle-Ottawa Scale for observational studies. Certainty of evidence was evaluated using GRADE (Supplemental Tables 4, 5).

Meta-Analysis and Data Synthesis

Meta-analysis was not feasible due to substantial heterogeneity. Study designs were predominantly case reports and case series with only 1 RCT. ¹ Treatment approaches varied widely (66 different regimens with variable dosing, duration, and routes). No standardized outcome measures were used, with complete resolution criteria varying across studies and follow-up ranging from 2 to 104.4 months. Clinical presentations varied (erosive, hypertrophic, mucosal variants) with differing disease severity and prior treatment exposure. We therefore conducted narrative synthesis stratified by treatment class and complexity.

Quality of Evidence Assessment

The overall quality of evidence was limited by the predominance of observational studies. Using the Oxford Centre for Evidence-Based Medicine levels, we identified: 1 Level 2 RCT (1.5%), ¹ 1 Level 2b prospective cohort study (1.5%), ² 16 Level 3 retrospective/multicenter studies (24.2%),^3-18 and 48 Level 4 case reports/series (72.7%).^1,19-65 This distribution reflects the rarity of VVLP and ethical challenges in randomizing patients with painful, erosive disease to placebo.

Study Characteristics

The initial search yielded 784 studies, with 442 duplicates removed using Covidence. In all, 66 studies meeting all inclusion criteria were included, encompassing 1417 patients with VVLP (Supplemental Figure 1). Studies excluded at full-text screening included cases only targeting oral lichen planus, conference abstracts, or not describing outcomes. Patient counts represent the total from included studies, assuming no overlap between study populations. Highest response rates were seen with systemic corticosteroids alone (88.9%, 24/27 patients), biologics with tildrakizumab (88%, 22/25 patients), and JAK inhibitors (75%, 6/8 patients) (Supplemental Tables 2, 3).

Reporting Bias Assessment/ GRADE Results

Among 66 included studies, only 1 RCT (Level 2) and 1 prospective cohort study (Level 2b) were identified, both with moderate RoB. Twelve retrospective studies (Level 3) had serious RoB (11/12), while 53 case reports/series (Level 4) formed the bulk of evidence. Using GRADE criteria, 94% (63/67) of studies were rated as very low certainty due to uncontrolled designs, small sample sizes, and lack of comparator groups. Only 4 studies achieved a low certainty rating (Supplemental Tables 4 and 5).

Due to study heterogeneity, we conducted descriptive synthesis rather than formal meta-analysis. Subgroup analysis showed higher response with intravaginal versus topical corticosteroids (55.7% vs 14.8%) and lower efficacy in patients with established scarring.

Treatment Approaches and Outcomes

Monotherapy (Single Drug Class) (25.2%, 357/1417 Patients)

All details about interventions, outcomes, and adverse effects can be found in Supplemental Table 3. Table 1 and Figure 1 summarize treatment outcomes by treatment class for monotherapy regimens (Figures 2 and 3).

OPEN IN VIEWER

Table 1.

Summary of Monotherapy Outcomes.

Treatment class	N	CR rate	PR rate	Recurrence	Key adverse events
Intravaginal CS suppositories	97	55.7% (54/97)	28.9% (28/97)	81.4% need maintenance	Minimal
Topical CS (clobetasol)	61	14.8% (9/61)	67.2% (41/61)	84%	Local irritation
Systemic CS (prednisolone)	27	88.9% (24/27)	7.4% (2/27)	Within weeks of taper	Cushingoid features (22.2%, 6/27)
Topical tacrolimus	42	26.2% (11/42)	50.0% (21/42)	0.25 m post-stop	Burning (52.4%, 22/42)
Methotrexate	27	0% (0/27)	100% (27/27)	2.75 m	Fatigue, GI distress, abnormal LFT (41%)
Hydroxychloroquine	15	60% (9/15)	20% (3/15)	NR	GI complaints, dizziness, infection (53.3%)
JAK inhibitors	8	75.0% (6/8)	25.0% (2/8)	2 m (1 patient)	ALT elevation, GI bleed (2/8)
Biologics (all)	31	71.0% (22/31)	29.0% (9/31)	3.2% at 23.8 m	Candidiasis (16.1%, 5/31)
Tildrakizumab only	25	88.0% (22/25)	12.0% (3/25)	3.2% at 23.8 m	As above
PDT	43	2.3% (1/43)	95.3% (41/43)	NR	Urinary soreness, bleeding
Isotretinoin	1	100% (1/1)	N/A	N/A	Lichenoid drug eruption (discontinued)

Abbreviations: ALT, alanine aminotransferase; CR, complete resolution; CS, corticosteroid; GI, gastrointestinal; HCQ, hydroxychloroquine; JAK, Janus kinase; LFTs, liver function tests; MTX, methotrexate; N/A, not applicable; NR, not reported; PDT, photodynamic therapy; PR, partial resolution.

OPEN IN VIEWER

Figure 1.

Distribution of treatment complexity across the study population.

OPEN IN VIEWER

Figure 2.

Monotherapy treatment outcomes in vulvovaginal lichen planus. CR, PR, and NOR rates are shown for each treatment class, ordered by CR rate. CR, complete resolution; PR, partial resolution; NOR, no response.

Intravaginal hydrocortisone acetate suppositories (25 mg) achieved 55.7% complete resolution (54 of 97 patients), nearly quadrupling the 14.8% (9 of 61 patients) seen with topical clobetasol propionate 0.05%.^19-26,46 The suppositories were typically applied 1 to 2 times daily. Suppositories were inserted intravaginally, while topical clobetasol was applied to vulvar surfaces using finger application.

Systemic Corticosteroids

Among 27 patients treated with prednisolone (typically 30 to 40 mg/day), 88.9% (24 of 27 patients) achieved complete resolution, but treatment carried significant toxicity, including Cushingoid features, osteopenia, and adrenal suppression (affecting 22.2%, 6 of 27 patients). Treatment durations ranged from 7.5 to 312 or more weeks.^26-28

Topical Tacrolimus

Among 42 patients treated with 0.1% ointment applied twice daily, 26.2% (11 of 42 patients) achieved CR, and 50.0% (21 of 42 patients) achieved PR, though burning occurred in 52.4% (22 of 42 patients). Those who tolerated it improved by 83.5% on average, though recurrence occurred rapidly (mean: 0.25 months).^29-31

Methotrexate

Among 27 patients treated with methotrexate (7.5-20 mg weekly, Cline 2020), all 27 (100%) achieved partial response, though none achieved complete resolution (0%). Adverse events included fatigue (6 patients), gastrointestinal distress (4 patients), and abnormal liver function tests (1 patient). Treatment discontinuation due to adverse events occurred in 8 of 27 patients (29.6%). Recurrence was reported at a mean interval of 2.75 months. ¹⁰

Hydroxychloroquine

Among 15 patients treated with hydroxychloroquine (200-600 mg/day, Vermeer 2021), 60.0% (9 of 15 patients) achieved complete resolution, and 20.0% (3 of 15 patients) achieved partial response. Median time to treatment response was 5 months. Adverse events occurred in 53.3% (8 of 15 patients), including gastrointestinal complaints, dizziness, and infection. One patient discontinued due to adverse events and 1 due to malignancy. ¹⁷

JAK Inhibitors

Among 8 patients, tofacitinib (5 mg BID, n = 7) achieved 85.7% complete resolution (6 of 7 patients), and abrocitinib (200 mg daily, n = 1) achieved partial resolution (95% lesion improvement). Overall, 75.0% (6 of 8 patients) achieved complete resolution, even in patients who had failed 8 to 10 prior treatments. Adverse events were limited to ALT elevation and GI bleeding (2 of 8 patients), and only 1 recurrence occurred after 2 months (Supplemental Tables 2 and 3).^35,36,59

Photodynamic Therapy

While only 2.3% (1/43) achieved complete resolution, 95.3% (41/43) had a partial response. Adverse events included urinary soreness (9 patients), bleeding (4 patients), and adhesion formation (2 patients).^1,63

Biologics

Among 31 patients treated with biologics as monotherapy, including tildrakizumab (25 patients) and adalimumab (6 patients), the overall CR rate was 71.0% (22 of 31 patients) with 88.2% mean improvement (Supplemental Tables 2 and 3). Tildrakizumab specifically achieved 88.0% CR (22 of 25 patients). The recurrence rate was the lowest among all treatments at 3.2% (1 of 31 patients), suggesting possible disease modification. Vulvovaginal candidiasis occurred in 16.1% (5 of 31 patients) (Supplemental Tables 2 and 3).^37,38,60

Retinoids

Isotretinoin (1 mg/kg/day) was reported in 1 patient as monotherapy over 12 weeks, achieving complete resolution (1 of 1 patient). However, treatment was subsequently discontinued due to a lichenoid drug eruption with cutaneous involvement, cheilitis, and dry eyes, raising concern about drug-induced exacerbation rather than therapeutic benefit. ⁵⁵

Adverse Events

Calcineurin inhibitors caused burning in 52.4% of patients but led to discontinuation in only 4.8%, suggesting increased tolerability over time with proper counseling.^{29 -31} Cases of refractory itch with the use of the medication may represent contact sensitization.

Systemic immunosuppressant infection rates were relatively low. Methotrexate caused gastrointestinal (GI) side effects (16.7%) and rare infections. Corticosteroids were associated with Cushingoid features and 1 treatment discontinuation due to concurrent malignancy (attribution to drug versus underlying disease process unclear).^16,28

Isotretinoin (1 mg/kg/day) led to treatment discontinuation due to lichenoid drug eruption with cutaneous involvement on the arms and legs, along with cheilitis and dry eyes, despite initially achieving complete resolution (see Monotherapy section).

Among 8 patients treated with tofacitinib (5 mg BID) or abrocitinib (200 mg daily), adverse events included ALT elevation (1 patient) and gastrointestinal bleeding leading to discontinuation (1 patient). ³⁵

In 31 patients treated with biologics (primarily tildrakizumab 200 mg monthly), vulvovaginal candidiasis occurred in 16.1% (5 patients), with mild injection-site erythema (1 patient) and moderate recurrent injection-site erythema leading to discontinuation (1 patient).^37,38

While most VVLP cases are idiopathic conditions, drug-induced lichenoid eruptions can involve the vulvovaginal area. Isotretinoin caused lichenoid drug reaction with cutaneous eruption (1 case), and 1 case report suggested metoprolol-induced VVLP with recurrence upon drug rechallenge.^42,55

Recurrence Patterns

Maintenance therapy was required in 81.4% of patients using intravaginal hydrocortisone acetate suppositories, with only 3% achieving durable remission. ⁵⁶

Time to recurrence varied considerably across treatment classes. Calcineurin inhibitors showed the most rapid recurrence, with disease returning within approximately 0.25 months of treatment discontinuation, essentially immediately upon cessation. ²⁹

Systemic corticosteroids similarly showed early recurrence, often within weeks of dose tapering.^21,26

JAK inhibitors demonstrated recurrence at approximately 2 months despite high initial efficacy, ³⁵ while methotrexate showed a somewhat longer interval of 2.75 months, suggesting a more gradual loss of disease control.^32,33

By contrast, biologics (tildrakizumab) had the longest recurrence-free interval at 23.8 months, with only 3.2% (1 of 31 patients) experiencing recurrence,^37,38 suggesting a potentially disease-modifying effect. Figure 3 illustrates the marked differences in time to recurrence across treatment classes.

OPEN IN VIEWER

Figure 3.

Time to recurrence by treatment class. Biologics (tildrakizumab) demonstrated the longest recurrence-free interval at 23.8 months.

These patterns suggest that VVLP may require indefinite maintenance therapy, similar to other chronic inflammatory conditions.

Treatment failure occurred in 17.5% of triple therapy and 32.1% of complex therapy patients. Most non-responders exhibited a treatment-resistant phenotype, having failed 5 or more prior therapies. Disease features associated with poor response included scarring, stenosis, and squamous cell carcinoma-related changes.^57,58 Intensive multi-agent regimens were linked to higher non-response rates. Adherence challenges with complex, multi-agent regimens may also contribute to treatment failure.

Intravaginal Versus Topical Corticosteroid Delivery

This review showed that intravaginal suppositories outperformed topical superpotent corticosteroids (CR 55.7% vs 14.8%).^19,20,46 This suggests mucosal drug delivery and contact time are important in addition to the class of corticosteroid potency in VVLP management.

This comparison is limited by the lack of clear reporting regarding anatomical disease distribution. The superior outcomes with intravaginal suppositories should be interpreted in the context of combined vulvovaginal disease, as most patients in these studies had both vaginal and vulvar involvement. Whether intravaginal suppositories would be effective for isolated vulvar disease without vaginal involvement remains unclear and warrants future investigation.

Long-term systemic corticosteroid use is associated with bone density loss. While intravaginal hydrocortisone suppositories had minimal reported adverse events in included studies, a theoretical risk of systemic absorption exists, particularly with prolonged use in inflamed mucosa. Rectal hydrocortisone formulations carry bone effect warnings; whether this applies to intravaginal administration remains unclear and warrants further investigation. ⁶⁶

Role of Combination Therapy

The widespread use of triple therapy (51.2%) reflects current clinical practice, though this must be interpreted carefully. The predominance of combination therapy may reflect a historical lack of highly effective single agents and the refractory nature of cases reaching specialized centers. The impressive efficacy of newer monotherapies suggests early use of targeted agents may reduce the need for complex regimens, though long-term comparative effectiveness remains to be established, alongside cost and accessibility issues.^43-47

Vaginal Dilators and Structural Management

The integration of vaginal dilators in 93.5% of dual therapy regimens acknowledges that VVLP’s morbidity stems not only from active inflammation but also from architectural distortion.^39-42 High partial response rates with mechanical therapy suggest that addressing both inflammatory and structural components is essential.

Standardization Needs

The heterogeneity in outcome reporting underscores the need for standardized measures in VVLP research. The C3 Chord organization has developed Core Outcome Sets for vulvovaginal diseases, including pain intensity, sexual function, psychological impact, and global impression of change using validated instruments. ⁶⁷ Adoption of these measures would enable meaningful study comparisons and meta-analyses. The absence of meta-analysis in our review reflects VVLP research limitations rather than methodological failure, highlighting the urgent need for standardized treatment protocols and comparative study designs.

Strengths and Limitations

This systematic review has several strengths, including a comprehensive search without restrictions, dual independent review, and transparent reporting of included and excluded studies.

Important limitations exist. The predominance of case reports and case series (72.7%) with only 1 RCT limits evidence strength. Study heterogeneity in design, treatments, outcomes, and populations prevented meta-analysis. Absent systematic gray literature searching may have missed studies, particularly unpublished negative findings, potentially overestimating efficacy. Publication bias may be less pronounced given the predominance of case reports versus RCTs. Most studies lacked control groups, standardized assessments, or adequate follow-up.

Lack of patient-level data prevented confounder adjustment, and overall sample size limited detection of rare adverse events. Sequential treatment patterns (novel agents in treatment-refractory patients, established therapies first line) prevent direct response rate comparisons.

While no language restrictions were applied to the search strategy, 7 non-English language studies were excluded at full-text review due to unavailable translation resources, which may have introduced language bias.

Future Research and Clinical Implications

Based on our findings, future research should investigate modified VVLP management approaches. Initial therapy with intravaginal corticosteroid suppositories (hydrocortisone acetate 25 mg) may be more effective than topical application, particularly with mechanical therapy (dilators), for patients with architectural changes. Studies should examine the role of non-steroidal topical alternatives such as JAK inhibitors or calcineurin inhibitors if tolerated in improving outcomes. Research is needed to determine optimal timing for systemic therapy (methotrexate 7.5-20 mg weekly or hydroxychloroquine 200-400 mg daily) in patients with inadequate topical response or extensive disease.

Priority should be given to RCTs comparing JAK inhibitors (tofacitinib 5 mg twice daily) and biologics (tildrakizumab 200 mg monthly) given their promising efficacy in treatment-refractory cases. Long-term maintenance studies are needed to identify minimum effective regimens for sustained control.

The identification of only 1 RCT among 66 studies underscores the need for high-quality comparative trials. Priority trials should compare intravaginal versus topical corticosteroid delivery, biologics and JAK inhibitors versus conventional systemic therapy in treatment-naïve patients, biologics versus JAK inhibitors for refractory disease, and continuous versus intermittent maintenance strategies.

The outcome heterogeneity that prevented quantitative synthesis necessitates the development of VVLP-specific disease activity scores, patient-reported outcome measures, standardized photographic protocols, and consensus treatment response definitions. Clinical or molecular predictors of treatment response would enable personalized therapy. Given that 51.2% receive triple therapy and novel agents are increasingly used, long-term safety registries are essential. Pragmatic trials addressing first-line therapy selection, treatment duration, and sequential algorithms would inform clinical practice.