Key Points
Question
What are the characteristics and factors associated with relapse after isotretinoin treatment for acne?
Findings
This cohort study including 19 907 patients found that 22.5% had acne relapse and 8.2% had isotretinoin retrial. Cumulative dosage was significantly associated with decreased rates of relapse and retrial (hazard ratio [HR], 0.99 for both); female sex was significantly associated with increased rate of relapse (HR, 1.43) and decreased rate of retrial (HR, 0.68); daily dose was not negatively associated with relapse or retrial among patients with cumulative dosages of 120 mg/kg or more.
Meaning
These findings suggest that higher cumulative dosages, but not daily doses, of isotretinoin are associated with lower rates of acne relapse and isotretinoin retrial.
Abstract
Importance
Isotretinoin is the only medical acne treatment capable of inducing acne remission; however, some patients experience acne relapse and require retrials of isotretinoin. There is a need to understand who is most at risk and how daily dose and cumulative dosage can influence outcomes.
Objective
To assess rates of acne relapse and isotretinoin retrial and to identify associated factors among patients with acne who received an isotretinoin treatment course.
Design, Setting, and Participants
This cohort study used data from the MarketScan commercial claims database from January 1, 2017, to December 31, 2020, to identify patients with acne who were 12 years or older and had received isotretinoin for 4 months or longer, with at least 1 year of continuous enrollment after completion of isotretinoin. Data analyses were performed from June 30, 2024, to August 1, 2024.
Main Outcomes and Measures
Multivariable Cox proportional hazards regression was used to quantify associations of patient demographic and treatment characteristics with acne relapse and isotretinoin retrial.
Results
A total of 19 907 patients (mean [SD] age, 20.6 [7.8] years; 10 504 females [52.8%]) were included, among whom 4482 (22.5%) had acne relapse and 1639 (8.2%) had isotretinoin retrial. Female sex (hazard ratio [HR], 1.43; 95% CI, 1.35-1.52) was significantly associated with increased rates of acne relapse, and isotretinoin cumulative dosage (mg/kg) was associated with a decreased rate of acne relapse (HR, 0.996; 95% CI, 0.995-0.997). Furthermore, daily dose was not associated with decreased risk of acne relapse or isotretinoin retrial among those with conventional and high cumulative dosages. Female sex (HR, 0.68; 95% CI, 0.62-0.76) and isotretinoin cumulative dosage (HR, 0.99; 95% CI, 0.98-0.99) were associated with decreased rates of isotretinoin retrial. Stratification by cumulative dosage indicated that higher cumulative dosage was associated with decreased rates of retrial among patients with low (<120 mg/kg) and conventional (120-220 mg/kg), but not high (>220 mg/kg) cumulative dosage. Maximum daily dose (mg/kg/d) was not negatively associated with acne relapse or isotretinoin retrial in patients with cumulative dosage of 120 mg/kg or more.
Conclusions and Relevance
The findings of this cohort study suggest that higher cumulative dosage may potentially reduce risk of acne relapse and isotretinoin retrial. Furthermore, daily dose was not associated with decreased risk of the outcomes for conventional and high cumulative dosage; therefore, daily dosing could be individualized to patient goals and preferences.
This cohort study assess rates of acne relapse and isotretinoin retrial and identifies associated factors in patients after completing an isotretinoin course.
Introduction
Isotretinoin is a highly effective treatment for severe acne and is the only medical treatment currently capable of providing lasting remission.1 Despite its efficacy, some patients experience acne relapse or require additional courses of isotretinoin. Given the dose-dependent adverse effects of isotretinoin,2 it is important to understand whether patient characteristics and treatment factors such as treatment regimen and dosage may be associated with the rate of acne relapse. This understanding can guide optimal isotretinoin regimens to balance adverse effects while maximizing the likelihood of achieving acne remission. Improved understanding may also aid in managing patient expectations and informing decisions on maintenance treatment plans.
Several observational studies3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19 have investigated the rates and predictive factors of acne relapse and isotretinoin retrial. These studies vary widely in their findings, with reports ranging from 9.4% to 65.4% for acne relapse and from 1.7% to 23.1% for isotretinoin retrial. This variability may be due to the use of clinical datasets limited by small sample size, unrepresentative populations, and incomplete follow-up (eg, patient may pursue care with a different clinician and not be captured). Moreover, inconsistent definitions of acne relapse and varying isotretinoin dosage regimens may further contribute to discrepancies in reported outcomes. These limitations create a knowledge gap on the rates and factors associated with acne relapse and isotretinoin retrial.
Claims data can overcome several of these limitations by providing a larger, more representative population, and mitigate follow-up bias by capturing prescriptions by any practitioner while patients are enrolled in the health plan.20,21 The purpose of this study was to assess the rates of acne relapse and isotretinoin retrial and identify factors associated with these outcomes in patients who completed a first course of isotretinoin.
Methods
Given the use of deidentified data, this cohort study was deemed exempt from review and informed consent by the Mass General Brigham Institutional Review Board. This study was reported in adherence with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Study Design and Data Source
We performed a retrospective cohort study using data from the MarketScan commercial claims database for the period from January 1, 2017, to December 31, 2020. This database includes deidentified commercial claims data for more than 1.85 million individuals with 1 or more claim(s) with an acne diagnosis. These longitudinal data included inpatient, outpatient, and pharmacy claims, as well as patient demographic information such as age and sex.
Study Population
Patients included in the analysis met the following inclusion criteria: had at least 1 encounter for a previously validated acne diagnosis per the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision code; were 12 years or older; had received a prescription for isotretinoin associated with an encounter for acne; had a first isotretinoin course duration of 4 months or more (to minimize inclusion of courses prematurely discontinued due to adverse effects or logistical barriers); and had at least 1 year of continuous enrollment after completion of isotretinoin (to ensure adequate follow-up time to identify acne relapse or isotretinoin retrial). The date of first isotretinoin course completion was defined as the index date. Prescriptions and doses for isotretinoin were identified using its US National Drug Code. Isotretinoin prescriptions separated by as many as 60 days between start dates were consolidated into the same course to account for possible nonadherence or logistical delays.
Outcomes
Outcomes were evaluated during at least 1 year of follow-up after the index date. Acne relapse was defined as receiving a prescription for a systemic acne treatment associated with a recent encounter for acne after the index date, including oral tetracycline antibiotics (ie, doxycycline, minocycline), other oral antibiotics commonly prescribed for acne (ie, trimethoprim-sulfamethoxazole, amoxicillin, cephalexin),22 spironolactone, and isotretinoin. Rates of topical acne prescriptions, including topical retinoids, antibiotics, benzoyl peroxide, and their fixed-dose combinations, were also evaluated. Prescriptions were identified using the US National Drug Code Directory. To avoid including oral antibiotic prescriptions for nonacne indications, only prescriptions for courses of more than 28 days were included.22 Isotretinoin retrial was defined as receiving another prescription for isotretinoin after the index date.
Covariates
Age at isotretinoin initiation (years), female sex, isotretinoin cumulative dosage (mg/kg), isotretinoin maximum daily dose (mg), isotretinoin treatment duration (months), and having a dermatologist practitioner were evaluated as potential factors associated with acne relapse and isotretinoin retrial. Having a dermatologist practitioner was defined as having a documented acne encounter with a dermatologist physician. Because weight is not available in the database, isotretinoin cumulative dosage (mg/kg) and maximum daily dose (mg/kg/d) were determined based on the average weight matched by the age and sex of the patient according to the 2021 US National Center for Health Statistics Anthropometric Reference Data.23 For stratified analyses, cumulative dosage was stratified into low (<120 mg/kg),13 conventional (120-220 mg/kg), and high (>220 mg/kg).15 Maximum daily doses were also stratified into low (<0.5 mg/kg/d), conventional (0.5-1.0 mg/kg/d), and high (>1 mg/kg/d) daily doses.24
Statistical Analysis
Univariate and multivariable Cox proportional hazards regression models were used to quantify associations between patient demographic and treatment characteristics and the outcomes (acne relapse and isotretinoin retrial) in patients treated with isotretinoin for acne with at least 1 year of follow-up. Variable inflation factors were calculated to select variables for multivariable analyses that avoid high multicollinearity (inflation factor <10). Primary multivariable analyses included age at isotretinoin initiation, female sex, isotretinoin cumulative dosage, and having a dermatologist practitioner. For stratified analyses, separate Cox proportional hazards regression models were fit for each stratum. Characteristics of patients with vs without acne relapse or isotretinoin retrial were compared using t test or χ2 test as appropriate. Tests were 2-tailed, and statistical significance was defined as P <. 05. Statistical analyses were performed using Python, version 3.8.2 (Python Software Foundation) from June 30, 2024, to August 1, 2024.
Results
Study Cohort
Among 1 856 012 patients with acne, 19 907 (mean [SD] age, 20.6 [7.8] years; 10 504 females [52.8%] and 9403 males [47.2%]) met the inclusion criteria for the study (Table 1). Most patients (87.0%) had an acne consultation with a dermatologist practitioner. Initial isotretinoin courses had a mean (SD) duration of 5.6 (1.6) months, maximum daily dose of 0.93 (0.27) mg/kg/d, and cumulative dosage of 132.4 (47.2) mg/kg. Mean (SD) follow-up time was 24.9 (8.7) months (Table 1).
Table 1. Characteristics of Patients With Acne Who Completed a Course of Isotretinoin With or Without Acne Relapse.
| Characteristic | No. (%) | P value | ||
|---|---|---|---|---|
| Total | No acne relapse | Acne relapse | ||
| Patients, No. | 19 907 | 15 425 (77.5) | 4482 (22.5) | NA |
| Age at isotretinoin initiation, mean (SD), y | 20.6 (7.8) | 20.5 (7.6) | 20.9 (8.5) | .008 |
| Female | 10 504 (52.8) | 7777 (50.4) | 2727 (60.8) | <.001 |
| Male | 9403 (47.2) | 7648 (49.6) | 1755 (39.2) | |
| Individuals with a dermatologist practitioner | 17 311 (87.0) | 13 330 (86.4) | 3981 (88.8) | <.001 |
| Isotretinoin treatment, mean (SD) | ||||
| Course duration, mo | 5.6 (1.6) | 5.6 (1.6) | 5.4 (1.6) | <.001 |
| Maximum daily dose, mg/kg/d | 0.93 (0.27) | 0.93 (0.26) | 0.92 (0.28) | .002 |
| Cumulative dosage, mg/kg | 132.4 (47.2) | 134.1 (46.8) | 126.3 (48.0) | <.001 |
| Follow-up time, mo | 24.9 (8.7) | 24.2 (8.6) | 27.5 (8.7) | <.001 |
| First acne systemic prescription after isotretinoin | ||||
| Time to prescription, median (IQR), mo | NA | NA | 7.5 (2.7-13.7) | NA |
| Prescription type | ||||
| Doxycycline | NA | NA | 954 (21.3) | NA |
| Minocycline | NA | NA | 660 (14.7) | NA |
| Other oral antibiotics | NA | NA | 386 (8.6) | NA |
| Spironolactone | NA | NA | 1006 (22.4) | NA |
| Isotretinoin retrial | NA | NA | 1476 (32.9) | NA |
| Months to acne relapse after isotretinoin | ||||
| <1 | NA | NA | 96 (2.1) | NA |
| 1 to 3 | NA | NA | 1155 (25.8) | NA |
| >3 to <6 | NA | NA | 699 (15.6) | NA |
| 6 to 12 | NA | NA | 1128 (25.2) | NA |
| >12 | NA | NA | 1404 (31.3) | NA |
Abbreviation: NA, not applicable.
Acne Relapse
Among the 19 907 patients in the cohort, 4482 (22.5%) had acne relapse (Table 1). There was a total of 34 638 person-years of follow-up, during which the rate of acne relapse was 12.9 per 100 person-years (Table 1). The median (IQR) time to acne relapse was 7.5 (2.7-13.7) months; 2.1% had a systemic acne prescription within 1 month of the index date; 25.8% within 1 to 3 months; 15.6% within 3 to 6 months; 25.2% within 6 to 12 months; and 31.3% after more than 12 months. Among patients with acne relapse, the most common systemic acne treatment prescribed after isotretinoin completion was oral antibiotics (doxycycline, 954 patients [21.3%]; minocycline, 660 [14.7%]; other oral antibiotics, 386 [8.6%]), followed by isotretinoin retrial (1476 [32.9%]) and spironolactone (1006 [22.4%]). In addition, 4228 patients (21.2%) received a topical acne prescription after completing isotretinoin (eTable 1 in Supplement 1). Sensitivity analyses including topical acne prescriptions identified 6960 patients (35%) who required subsequent topical or systemic acne treatment.
Univariate analyses are summarized in eTable 2 in Supplement 1. In multivariable Cox proportional hazards regression modeling, female sex (hazard ratio [HR], 1.43; 95% CI, 1.35-1.52) and having a dermatologist practitioner (HR, 1.21; 95% CI, 1.11-1.33) were significantly associated with increased rates of acne relapse, and isotretinoin cumulative dosage was significantly associated with a decreased rate of acne relapse (HR, 0.996; 95% CI, 0.995-0.997; Table 2). There was no evidence of an interaction between age and sex (eTable 3 in Supplement 1). After stratification into adolescents (age <18 years) and adults (age ≥18 years), female sex was still significantly associated with increased rates of acne relapse for both age groups (eTable 4 in Supplement 1).
Table 2. Multivariable Cox Proportional Hazards Regression Models for Acne Relapse and Isotretinoin Retrial.
| Factor | Acne relapse | Isotretinoin retrial | ||
|---|---|---|---|---|
| Adjusted HR (95% CI) | P value | Adjusted HR (95% CI) | P value | |
| Age at isotretinoin initiation, y | 1.00 (0.99-1.00) | .72 | 0.98 (0.98-0.99) | <.001 |
| Female sex | 1.43 (1.35-1.52) | <.001 | 0.68 (0.62-0.76) | <.001 |
| Isotretinoin cumulative dosage, mg/kg | 0.996 (0.995-0.997) | <.001 | 0.99 (0.98-0.99) | <.001 |
| Dermatologist practitioner | 1.21 (1.11-1.33) | <.001 | 1.17 (1.00-1.36) | .04 |
Abbreviation: HR, hazard ratio.
Maximum daily dose and isotretinoin cumulative dosage could not be simultaneously included in the model due to high variable inflation factors.
Isotretinoin Retrial
Among the 19 907 patients in the cohort, 1639 (8.2%) underwent isotretinoin retrial (1454 had 2 courses, 144 had 3 courses, 35 had 4 courses, and 6 had 5 or more courses; eTable 5 in Supplement 1). There was a total of 38 068 person-years of follow-up, with an isotretinoin retrial rate of 4.3 per 100 person-years. The median (IQR) time to a second isotretinoin course was 2.8 (2.2-6.5) months. Mean (SD) second isotretinoin course duration was 2.3 (1.9) months; maximum daily dose was 0.82 (0.30) mg/kg/d; and cumulative dosage was 52.6 (47.1) mg/kg (eTable 5 in Supplement 1). Among patients with isotretinoin retrial, 364 (22.2%) had a second isotretinoin course of 4 months or longer duration, and 437 (26.7%) had a retrial at least 6 months after completing the initial course. Retrial cumulative dosage and duration were significantly higher in patients with retrial after 6 months compared to patients with retrial within 6 months, and there was no significant difference in maximum daily dose (eFigure in Supplement 1).
Multivariable modeling demonstrated that age (HR, 0.98; 95% CI, 0.98-0.99), female sex (HR, 0.68; 95% CI, 0.62-0.76), and cumulative dosage (HR, 0.99; 95% CI, 0.98-0.99) were associated with decreased rates of isotretinoin retrial, and that having a dermatologist practitioner (HR, 1.17; 95% CI, 1.00-1.36) was associated with an increased rate of isotretinoin retrial (Table 2). There was evidence of an interaction between age and sex for isotretinoin retrial (HR, 1.02; 95% CI, 1.00-1.03), indicating that female sex was associated with less of a protective effect from increasing age compared to males (eTable 3 in Supplement 1). Stratified analyses of adolescents and adults showed that female sex was still significantly associated with decreased rates of isotretinoin retrial for both age groups (eTable 4 in Supplement 1).
Stratification by Isotretinoin Cumulative Dosage and Maximum Daily Dose
After stratification by cumulative dosage, higher cumulative dosage was significantly associated with decreased rates of acne relapse among patients within the low (HR, 0.990; 95% CI, 0.989-0.993) cumulative dosage group, but not the conventional or high cumulative dosage group (Table 3). Maximum daily dose was protective against acne relapse in the low cumulative dosage group (HR, 0.80; 95% CI, 0.65-0.99), associated with an increased rate of acne relapse in the conventional cumulative dosage group (HR, 1.41; 95% CI, 1.13-1.77), and not a significant estimator for the high cumulative dosage group (eTable 6 in Supplement 1). Female sex remained significantly associated with increased rates of acne relapse for all cumulative dosage groups (Table 3).
Table 3. Multivariable Cox Proportional Hazards Regression Models for Acne Relapse Stratified by Cumulative Dosage.
| Factor | Acne relapse, cumulative dosage, mg/kg | |||||
|---|---|---|---|---|---|---|
| Low (<120) | Conventional (120-220) | High (>220) | ||||
| Adjusted HR (95% CI) | P value | Adjusted HR (95% CI) | P value | Adjusted HR (95% CI) | P value | |
| Patients, No. | 8328 | NA | 10 778 | NA | 801 | NA |
| Age at isotretinoin initiation, y | 1.00 (0.99-1.00) | .55 | 1.00 (0.99-1.00) | .61 | 1.00 (0.97-1.02) | .85 |
| Female sex | 1.15 (1.05-1.25) | .002 | 1.80 (1.64-1.97) | <.001 | 1.77 (1.28-2.43) | <.001 |
| Isotretinoin cumulative dosage | 0.990 (0.989-0.993) | <.001 | 1.00 (0.998-1.00) | .94 | 0.997 (0.993-1.001) | .17 |
| Dermatologist practitioner | 1.18 (1.03-1.34) | .01 | 1.22 (1.06-1.40) | <.001 | 2.22 (1.21-4.09) | .01 |
Abbreviations: HR, hazard ratio; NA, not applicable.
Maximum daily dose and isotretinoin cumulative dosage could not be included simultaneously in the model due to high variable inflation factors.
Higher cumulative dosage was significantly associated with decreased rates of isotretinoin retrial in the low (HR, 0.98; 95% CI, 0.98-0.99) and conventional (HR, 0.993; 95% CI, 0.989-0.996) cumulative dosage groups. However, cumulative dosage was not significantly associated with isotretinoin retrial among patients in the high cumulative dosage group (Table 4). Maximum daily dose was significantly associated with decreased rates of isotretinoin retrial for the low cumulative dosage group (HR, 0.59; 95% CI, 0.44-0.79) and increased rates of isotretinoin retrial for the conventional cumulative dosage group (HR, 1.61; 95% CI, 1.04-2.51; eTable 7 in Supplement 1). Female sex was significantly associated with decreased rates of isotretinoin retrial among patients with low (HR, 0.66; 95% CI, 0.59-0.75) and conventional (HR, 0.74; 95% CI, 0.62-0.88) cumulative dosage (Table 4).
Table 4. Multivariable Cox Proportional Hazards Regression Models for Isotretinoin Retrial Stratified by Cumulative Dosage.
| Factor | Isotretinoin retrial, cumulative dosage, mg/kg | |||||
|---|---|---|---|---|---|---|
| Low (<120) | Conventional (120-220) | High (>220) | ||||
| Adjusted HR (95% CI) | P value | Adjusted HR (95% CI) | P value | Adjusted HR (95% CI) | P value | |
| Patients, No. | 8328 | NA | 10 778 | NA | 801 | NA |
| Age at isotretinoin initiation, y | 0.99 (0.98-0.99) | <.001 | 0.97 (0.95-0.98) | <.001 | 0.96 (0.90-1.02) | .16 |
| Female sex | 0.66 (0.59-0.75) | <.001 | 0.74 (0.62-0.88) | <.001 | 0.71 (0.39-1.29) | .26 |
| Isotretinoin cumulative dosage, mg/kg | 0.98 (0.98-0.99) | <.001 | 0.993 (0.989-0.996) | <.001 | 1.00 (0.99-1.00) | .58 |
| Dermatologist practitioner | 1.05 (0.88-1.26) | .55 | 1.39 (1.05-1.85) | .02 | 6.57 (0.90-47.75) | .06 |
Abbreviations: HR, hazard ratio; NA, not applicable.
Maximum daily dose and isotretinoin cumulative dosage could not be included simultaneously in the model due to high variable inflation factors.
Patients with conventional and high maximum daily doses of isotretinoin had lower rates of acne relapse (22.6% and 21.7%, respectively) compared to low maximum daily doses (30.2%; Table 5). They also had lower rates of isotretinoin retrial (conventional, 8.5%; high, 7.1%) compared to low maximum daily doses (16.2%). Similarly, patients with conventional and high cumulative dosages of isotretinoin had lower rates of acne relapse (low, 26.1%; conventional, 19.8%; high, 22.2%) and isotretinoin retrial (low, 12.7%; conventional, 5.0%; high, 5.5%) compared to those with low cumulative dosages (Table 5).
Table 5. Characteristics of Patients With Acne, Stratified by Isotretinoin Daily and Cumulative Dosage.
| Characteristic | Isotretinoin treatment | |||||
|---|---|---|---|---|---|---|
| Maximum daily, mg/kg/d | Cumulative dosage, mg/kg | |||||
| Low (<0.5) | Conventional (0.5-1.0) | High (>1.0) | Low (<120) | Conventional (120-220) | High (>220) | |
| Patients, No. | 777 | 10 527 | 8603 | 8328 | 10 778 | 801 |
| Age at isotretinoin initiation, mean (SD), y | 27.3 (12.0) | 21.1 (8.1) | 19.4 (6.5) | 21.6 (8.6) | 20.0 (7.2) | 18.8 (7.0) |
| Female, No. (%) | 383 (49.3) | 5649 (53.7) | 4472 (52.0) | 4262 (51.2) | 5786 (53.7) | 456 (56.9) |
| Male, No. (%) | 394 (50.7) | 4878 (46.3) | 4131 (48.0) | 4066 (48.8) | 4992 (46.3) | 345 (43.1) |
| Individuals with a dermatologist practitioner, No. (%) | 653 (84.0) | 9205 (87.4) | 7453 (86.6) | 7209 (86.6) | 9401 (87.2) | 701 (87.5) |
| Isotretinoin course duration, mean (SD), mo | 6.0 (2.3) | 5.6 (1.7) | 5.5 (1.5) | 4.9 (1.2) | 5.9 (1.5) | 8.3 (2.6) |
| Isotretinoin maximum daily dose, mean (SD), mg/kg/d | 0.4 (0.09) | 0.78 (0.14) | 1.2 (0.18) | 0.76 (0.22) | 1.0 (0.21) | 1.3 (0.33) |
| Isotretinoin cumulative dosage, mean (SD), mg/kg | 65.4 (29.1) | 115.1 (34.5) | 159.6 (45.9) | 91.7 (20.8) | 154.3 (24.8) | 259.4 (52.3) |
| Individuals with acne relapse, No. (%) | 235 (30.2) | 2382 (22.6) | 1865 (21.7) | 2175 (26.1) | 2129 (19.8) | 178 (22.2) |
| Time to acne relapse, mean (SD), mo | 7.5 (7.1) | 9.3 (7.8) | 10.0 (8.1) | 8.4 (7.8) | 10.6 (7.9) | 9.5 (7.8) |
| Individuals with isotretinoin retrial, No. (%) | 126 (16.2) | 898 (8.5) | 615 (7.1) | 1056 (12.7) | 539 (5.0) | 44 (5.5) |
| Time to second isotretinoin course, mean (SD), mo | 4.1 (3.3) | 5.2 (4.7) | 5.4 (5.0) | 4.5 (4.3) | 6.4 (5.3) | 6.7 (5.1) |
| Follow-up time, mean (SD), mo | 24.7 (8.7) | 24.7 (8.8) | 25.2 (8.7) | 25.4 (9.1) | 24.7 (8.4) | 23.6 (7.7) |
Discussion
This cohort study of 19 907 patients with acne treated with isotretinoin found that 22.5% experienced acne relapse requiring further treatment with systemic acne medications, and 8.2% underwent a retrial of isotretinoin. Higher isotretinoin cumulative dosage was associated with reduced rates of acne relapse and isotretinoin retrial, corroborating the findings of prior smaller observational studies.10,14,15,19,25 When stratified by cumulative dosage, higher maximum daily dose was only associated with reduced rates of acne relapse and isotretinoin retrial for the low cumulative dosage group. Within the conventional cumulative dosage group, higher maximum daily dose was associated with increased rates of acne relapse and isotretinoin retrial. The underlying factors for this association are uncertain, and could include increased adverse effects from higher daily dose leading to earlier course termination, prescriber preferences, or confounding related to disease severity.26 Moreover, even without considering cumulative dosage, the rate of acne relapse was comparable between conventional (0.5-1.0 mg/kg/d) and high (>1.0 mg/kg/d) daily doses (22.6% vs 21.7%). Prior studies2,15,27 have demonstrated increased risk of adverse effects with higher daily doses of isotretinoin. Our study suggests that lower daily doses may still achieve similar rates of acne remission if a sufficient cumulative dosage is reached. Overall, these findings suggest that daily dosing can be individualized to patient goals and preferences, balancing course duration and adverse effects. Although cumulative dosage was significantly associated with a decreased rate of acne relapse in the low cumulative dosage group and a decreased rate of isotretinoin retrial in the low and conventional cumulative dosage groups, it was not significantly associated with either outcome in the high cumulative dosage group. This suggests that beyond 220 mg/kg, higher cumulative dosage may not be helpful for preventing acne relapse or the need for another course of isotretinoin.
Second courses of isotretinoin had an average cumulative dosage less than 120 mg/kg, suggesting that retrials tended to be partial courses. This may be due to decreased acne severity at the time of retrial, thus requiring a shorter course, or early discontinuation due to intolerable adverse effects or logistical barriers.28 Stratification by timing of isotretinoin retrial revealed that patients with isotretinoin retrial within 6 months had significantly lower retrial cumulative dosage and course duration than those with retrial after 6 months. These findings further suggest that early retrial courses may represent a continuation of the initial course due to a break in therapy due to substantial logistical barriers, or those with early retrial may reach acne clearance faster, potentially due to differences in acne severity at the time of retrial, or residual benefits from the initial course.
In this study, females were significantly more likely than males to experience recurrence of acne, and this trend remained without any interaction between age and sex. This suggests that it is more difficult for females to remain clear from acne following a course of isotretinoin. Moreover, while females and increased age were associated with a decreased rate of isotretinoin retrial, there was significant interaction between age and sex, indicating that the protective effect of age is weaker in females compared to males. These findings are consistent with prior evidence29 indicating that females tend to have more persistent acne than males into adulthood. Several factors could contribute to this finding, including comorbidities such as polycystic ovarian syndrome and hyperandrogenism,30,31 and increased rates of visiting a dermatologist practitioner for acne seen in female patients.32 Additionally, patient preferences and a lower threshold for treating recurrent lesions with systemic acne therapies among females compared to males may influence measured rates of acne relapse.
Our study found that female patients are less likely than males to undergo a retrial of isotretinoin; however, in 2011 Coloe et al found that females had an increased risk of isotretinoin retrial.14 This discrepancy may be attributed to the increased prescribing of spironolactone for female patients with acne in recent years.22 In our study, among the 1006 patients who had a prescription for spironolactone after completing isotretinoin, only 41 had a subsequent isotretinoin retrial. This finding suggests that the option to use spironolactone may be associated with a reduced likelihood of isotretinoin retrial among female patients with acne relapse. In addition, logistical barriers due to iPledge28 requirements may contribute to decreased rates of isotretinoin retrial in females compared to males.
Prior to our study, the largest study on isotretinoin outcomes in patients with acne was conducted by Azoulay et al10 in 2007; it included 17 351 patients from 1984 to 2003. Similar to their findings, our study identified that increased isotretinoin cumulative dosage and longer course duration were associated with a decreased risk of acne relapse and isotretinoin retrial. In addition, both studies found that having a dermatologist was associated with an increased risk of acne relapse. This may reflect greater acne severity or more frequent acne follow-up among patients who see a dermatologist. It may also reflect that dermatologists are more likely to prescribe systemic acne treatments and retrial isotretinoin in patients with acne recurrence after the first course of isotretinoin. However, there are several notable differences between these studies. Compared to our study, Azoulay et al10 reported higher rates of acne relapse (41% vs 22.5%) and isotretinoin retrial (26% vs 8.2%). This may partially be explained by their broader definition of acne relapse as receiving a topical or systemic acne medication. In our cohort, when we additionally included topical medications in our definition of relapse, we found that 35% of patients received a topical and/or systemic acne medication after isotretinoin. However, considering that topical therapy may be used for prevention or spot treatment rather than for true acne relapse, it is important to interpret these findings cautiously. Additionally, the cohort in the Azoulay et al study10 had a follow-up time of up to 20 years, potentially contributing to higher observed rates of acne relapse and isotretinoin retrial. Unlike our study, they did not require their cohort to meet a minimum isotretinoin course duration. Therefore, their cohort10 may have had higher rates of relapse or isotretinoin retrial if patients had an incomplete initial course of isotretinoin. By requiring that patients complete at least 4 months of isotretinoin as part of our inclusion criteria, we assessed rates of acne relapse and isotretinoin retrial among patients who were more likely to have completed a full course of isotretinoin. Lastly, their study10 was based on a Canadian population and they found that the publishing of the Canadian acne guidelines influenced the rate of isotretinoin retrial. Differences in acne treatment guidelines over time and across countries may also account for the varying results.
Limitations
An important limitation of this study was the absence of data on patient weight in the MarketScan database. To address this limitation, weight-based daily dose and cumulative dosage were estimated using average weight data from the 2021 US National Center for Health Statistics report matched by age and sex.23 Although these reference data were the most relevant to our study period, an important limitation is that the reference data were from 2015 to 2018, whereas our study period was 2017 to 2020. Another limitation is that acne relapse was defined as receiving a prescription for a systemic acne medication; however, the indication for the prescription could be treatment switching, maintenance, or relapse. However, most individuals with a systemic acne prescription after isotretinoin received it more than 1 month later (97.9%), suggesting that these prescriptions were for acne relapse. In addition, although the systemic acne prescriptions were associated with a recent encounter for acne, it is possible that they were prescribed for conditions other than acne, such as longer courses of antibiotics for rosacea or hidradenitis suppurativa, and spironolactone for hypertension and androgenetic alopecia. Also, the absence of clinical data on acne severity at the time of isotretinoin initiation and discontinuation were limiting factors because higher cumulative dosage may be correlated with acne clearance, and individuals should be treated until they reach clinical remission rather than a predetermined cumulative dosage.33 Another limitation was that we could not account for potential previous courses of isotretinoin that were not captured in the MarketScan database. There may be additional factors associated with acne relapse and isotretinoin retrial that were not considered in this study, eg, acne severity, distribution of acne, and use of prescription and over-the-counter topical medications. Lastly, because MarketScan is largely based on commercial claims data, future studies are needed to evaluate whether these findings generalize to patients insured by Medicaid or Medicare plans.
Conclusions
This large cohort study identified factors associated with acne relapse and isotretinoin retrial that may guide treatment selection and prognosis. Further research is needed to understand optimal strategies to prevent acne relapse among those at highest risk.
eTable 1. Characteristics of topical acne prescription use after completion of isotretinoin.
eTable 2. Univariate Cox proportional hazards regression models for acne relapse and isotretinoin retrial.
eTable 3. Multivariable Cox proportional hazards regression models for acne relapse and isotretinoin retrial adjusting for interaction between age and sex.
eTable 4. Age stratified multivariable Cox proportional hazards regression models for acne relapse and isotretinoin retrial
eTable 5. Characteristics of acne patients treated with single versus multiple courses of isotretinoin.
eTable 6. Multivariable Cox proportional hazards regression models for acne relapse stratified by cumulative dosage.
eTable 7. Multivariable Cox proportional hazards regression models for isotretinoin retrial stratified by cumulative dosage.
eFigure. Characteristics of second isotretinoin course in relation to timing of retrial.
Data Sharing Statement
References
- 1.Reynolds RV, Yeung H, Cheng CE, et al. Guidelines of care for the management of acne vulgaris. J Am Acad Dermatol. 2024;90(5):1006.e1-1006.e30. doi: 10.1016/j.jaad.2023.12.017 [DOI] [PubMed] [Google Scholar]
- 2.Rademaker M. Adverse effects of isotretinoin: a retrospective review of 1743 patients started on isotretinoin. Australas J Dermatol. 2010;51(4):248-253. doi: 10.1111/j.1440-0960.2010.00657.x [DOI] [PubMed] [Google Scholar]
- 3.Jones DH, Cunliffe WJ. Remission rates in acne patients treated with various doses of 13-cis-retinoic acid (isotretinoin). Br J Dermatol. 1984;111(1):123-125. doi: 10.1111/j.1365-2133.1984.tb04030.x [DOI] [PubMed] [Google Scholar]
- 4.Stainforth JM, Layton AM, Taylor JP, Cunliffe WJ. Isotretinoin for the treatment of acne vulgaris: which factors may predict the need for more than one course? Br J Dermatol. 1993;129(3):297-301. doi: 10.1111/j.1365-2133.1993.tb11850.x [DOI] [PubMed] [Google Scholar]
- 5.Layton AM, Knaggs H, Taylor J, Cunliffe WJ. Isotretinoin for acne vulgaris–10 years later: a safe and successful treatment. Br J Dermatol. 1993;129(3):292-296. doi: 10.1111/j.1365-2133.1993.tb11849.x [DOI] [PubMed] [Google Scholar]
- 6.White GM, Chen W, Yao J, Wolde-Tsadik G. Recurrence rates after the first course of isotretinoin. Arch Dermatol. 1998;134(3):376-378. doi: 10.1001/archderm.134.3.376 [DOI] [PubMed] [Google Scholar]
- 7.Lehucher-Ceyrac D, de La Salmonière P, Chastang C, Morel P. Predictive factors for failure of isotretinoin treatment in acne patients: results from a cohort of 237 patients. Dermatology. 1999;198(3):278-283. doi: 10.1159/000018130 [DOI] [PubMed] [Google Scholar]
- 8.Al-Mutairi N, Manchanda Y, Nour-Eldin O, Sultan A. Isotretinoin in acne vulgaris: a prospective analysis of 160 cases from Kuwait. J Drugs Dermatol. 2005;4(3):369-373. [PubMed] [Google Scholar]
- 9.Quéreux G, Volteau C, N’Guyen JM, Dréno B. Prospective study of risk factors of relapse after treatment of acne with oral isotretinoin. Dermatology. 2006;212(2):168-176. doi: 10.1159/000090658 [DOI] [PubMed] [Google Scholar]
- 10.Azoulay L, Oraichi D, Bérard A. Isotretinoin therapy and the incidence of acne relapse: a nested case-control study. Br J Dermatol. 2007;157(6):1240-1248. doi: 10.1111/j.1365-2133.2007.08250.x [DOI] [PubMed] [Google Scholar]
- 11.Liu A, Yang DJ, Gerhardstein PC, Hsu S. Relapse of acne following isotretinoin treatment: a retrospective study of 405 patients. J Drugs Dermatol. 2008;7(10):963-966. [PubMed] [Google Scholar]
- 12.Harms M, Masouyé I, Radeff B. The relapses of cystic acne after isotretinoin treatment are age-related: a long-term follow-up study. Dermatologica. 1986;172(3):148-153. doi: 10.1159/000249320 [DOI] [PubMed] [Google Scholar]
- 13.Borghi A, Mantovani L, Minghetti S, Giari S, Virgili A, Bettoli V. Low-cumulative dose isotretinoin treatment in mild-to-moderate acne: efficacy in achieving stable remission. J Eur Acad Dermatol Venereol. 2011;25(9):1094-1098. doi: 10.1111/j.1468-3083.2010.03933.x [DOI] [PubMed] [Google Scholar]
- 14.Coloe J, Du H, Morrell DS. Could higher doses of isotretinoin reduce the frequency of treatment failure in patients with acne? J Am Acad Dermatol. 2011;65(2):422-423. doi: 10.1016/j.jaad.2010.06.025 [DOI] [PubMed] [Google Scholar]
- 15.Blasiak RC, Stamey CR, Burkhart CN, Lugo-Somolinos A, Morrell DS. High-dose isotretinoin treatment and the rate of retrial, relapse, and adverse effects in patients with acne vulgaris. JAMA Dermatol. 2013;149(12):1392-1398. doi: 10.1001/jamadermatol.2013.6746 [DOI] [PubMed] [Google Scholar]
- 16.Morales-Cardona CA, Sánchez-Vanegas G. Acne relapse rate and predictors of relapse following treatment with oral isotretinoin. Actas Dermosifiliogr. 2013;104(1):61-66. doi: 10.1016/j.ad.2012.05.004 [DOI] [PubMed] [Google Scholar]
- 17.Fernandez J, Lee B, Patel JM, et al. Retrospective case series of isotretinoin outcomes for acne in 393 female patients at Baylor College of Medicine during 2012-2016. J Am Acad Dermatol. 2020;82(5):1218-1219. doi: 10.1016/j.jaad.2019.06.1293 [DOI] [PubMed] [Google Scholar]
- 18.Demirci Saadet E. Investigation of relapse rate and factors affecting relapse after oral isotretinoin treatment in patients with acne vulgaris. Dermatol Ther. 2021;34(6):e15109. doi: 10.1111/dth.15109 [DOI] [PubMed] [Google Scholar]
- 19.Skroza N, Tolino E, Balduzzi V, et al. Advantages of tailored isotretinoin treatment in moderate to severe acne: real-life data. Front Pharmacol. 2021;12:733526. doi: 10.3389/fphar.2021.733526 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kulaylat AS, Schaefer EW, Messaris E, Hollenbeak CS. Truven Health Analytics MarketScan databases for clinical research in colon and rectal surgery. Clin Colon Rectal Surg. 2019;32(1):54-60. doi: 10.1055/s-0038-1673354 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Tripathi R, Mazmudar RS, Knusel KD, Bordeaux JS, Scott JF. Big data in dermatology: publicly available health care databases for population health research. J Am Acad Dermatol. 2020;83(5):1546-1556. doi: 10.1016/j.jaad.2020.04.145 [DOI] [PubMed] [Google Scholar]
- 22.Barbieri JS. Temporal trends in the use of systemic medications for acne from 2017 to 2020. JAMA Dermatol. 2023;159(10):1135-1136. doi: 10.1001/jamadermatol.2023.2363 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Fryar CD, Carroll MD, Gu Q, Afful J, Ogden CL. Anthropometric reference data for children and adults: United States, 2015-2018. Vital Health Stat 3. 2021;(36):1-44. [PubMed] [Google Scholar]
- 24.Rao PK, Bhat RM, Nandakishore B, Dandakeri S, Martis J, Kamath GH. Safety and efficacy of low-dose isotretinoin in the treatment of moderate to severe acne vulgaris. Indian J Dermatol. 2014;59(3):316. doi: 10.4103/0019-5154.131455 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Rademaker M. Isotretinoin: dose, duration and relapse. What does 30 years of usage tell us? Australas J Dermatol. 2013;54(3):157-162. doi: 10.1111/j.1440-0960.2012.00947.x [DOI] [PubMed] [Google Scholar]
- 26.Daly AU, Baptista Gonçalves R, Lau E, et al. A systematic review of isotretinoin dosing in acne vulgaris. JEADV Clin Pract. 2023;2(3):432-449. doi: 10.1002/jvc2.154 [DOI] [Google Scholar]
- 27.Lee JW, Yoo KH, Park KY, et al. Effectiveness of conventional, low-dose and intermittent oral isotretinoin in the treatment of acne: a randomized, controlled comparative study. Br J Dermatol. 2011;164(6):1369-1375. doi: 10.1111/j.1365-2133.2010.10152.x [DOI] [PubMed] [Google Scholar]
- 28.Charrow A, Xia FD, Lu J, Waul M, Joyce C, Mostaghimi A. Differences in isotretinoin start, interruption, and early termination across race and sex in the iPLEDGE era. PLoS One. 2019;14(3):e0210445. doi: 10.1371/journal.pone.0210445 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Collier CN, Harper JC, Cafardi JA, et al. The prevalence of acne in adults 20 years and older. J Am Acad Dermatol. 2008;58(1):56-59. doi: 10.1016/j.jaad.2007.06.045 [DOI] [PubMed] [Google Scholar]
- 30.Lehucher-Ceyrac D, Chaspoux C, Weber MJ, Morel P, Vexiau P. [Acne, hyperandrogenism and oral isotretinoin resistance. 23 cases. Therapeutic implications]. Ann Dermatol Venereol. 1997;124(10):692-695. [PubMed] [Google Scholar]
- 31.Cakir GA, Erdogan FG, Gurler A. Isotretinoin treatment in nodulocystic acne with and without polycystic ovary syndrome: efficacy and determinants of relapse. Int J Dermatol. 2013;52(3):371-376. doi: 10.1111/j.1365-4632.2012.05691.x [DOI] [PubMed] [Google Scholar]
- 32.Chang J, Nock MR, Cohen JM, Bunick CG. Acne accounts for an almost 2.5-fold higher proportion of dermatology visits among adult females compared to adult males in the United States: a study of the national ambulatory medical care survey from 2002-2016. PLoS One. 2023;18(9):e0290763. doi: 10.1371/journal.pone.0290763 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Rademaker M. Making sense of the effects of the cumulative dose of isotretinoin in acne vulgaris. Int J Dermatol. 2016;55(5):518-523. doi: 10.1111/ijd.12942 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Characteristics of topical acne prescription use after completion of isotretinoin.
eTable 2. Univariate Cox proportional hazards regression models for acne relapse and isotretinoin retrial.
eTable 3. Multivariable Cox proportional hazards regression models for acne relapse and isotretinoin retrial adjusting for interaction between age and sex.
eTable 4. Age stratified multivariable Cox proportional hazards regression models for acne relapse and isotretinoin retrial
eTable 5. Characteristics of acne patients treated with single versus multiple courses of isotretinoin.
eTable 6. Multivariable Cox proportional hazards regression models for acne relapse stratified by cumulative dosage.
eTable 7. Multivariable Cox proportional hazards regression models for isotretinoin retrial stratified by cumulative dosage.
eFigure. Characteristics of second isotretinoin course in relation to timing of retrial.
Data Sharing Statement
