Mechanism of Action

Isotretinoin, the cis-isomer of transretinoic acid, is converted in vivo to all-trans retinoic acid. The latter is the active effector molecule inducing cellular effects via binding to nuclear retinoic acid receptors, retinoid X receptors (RXRs), and retinoid acid receptors (RARs). The pathogenic mechanisms of acne involve sebum hypersecretion, intraductal epithelial hyperkeratinization, Propionibacterium acnes proliferation, and inflammation. Although isotretinoin has been shown to affect these pathogenic factors, only recently have the mechanisms been understood.

Isotretinoin has been shown to have direct effects on sebocytes, leading to sebum suppression, and on inflammation by inhibiting innate immune system activation. In sebocytes, isotretinoin induces cell cycle arrest and apoptosis by a mechanism independent of RAR binding. This effect likely contributes to its sebosuppressive activity and ameliorative effect on acne. Sebocyte apoptosis was shown to be mediated by tumor necrosis factor–related apoptosis-inducing ligand and neutrophil gelatinase-associated lipocalin (NGAL). The latter functions in innate immune defense against gram-negative bacteria. In patients with acne treated with isotretinoin, increased skin surface NGAL levels have been observed; an effect that precedes sebum suppression and reduction in P acnes . It is speculated that reduction in the level of P acnes may be caused by the combined effect of these outcomes.

Peripheral blood monocytes from patients with acne have been shown to express higher levels of toll-like receptor 2 (TLR-2) and show greater expression of TLR-2 following exposure to P acnes . Isotretinoin at 1 week reduced monocyte TLR-2 expression and inflammatory cytokine release induced by P acnes . Note that this effect persisted to 6 months posttreatment, implying that TLR-2 modulation may be involved in the long-term therapeutic response to isotretinoin.

Recently, oral isotretinoin was found to result in significant changes in various hormones, including some implicated in acne pathogenesis, such as insulin-like growth factor 1 and growth hormone, in a dose-dependent manner. In 105 patients with acne divided into 3 isotretinoin dosing groups (0.5–1 mg/kg/d, 0.2–0.5 mg/kg/d, and intermittent 0.5–1 mg/kg/d for 1 week per month) hormone levels were measured at 3 months. Levels of luteinizing hormone, prolactin, total testosterone, adrenocorticotropic hormone, cortisol, insulin-like growth factor 1, growth hormone, and free T3 and T4 were reduced, whereas levels of dehydroepiandrosterone sulfate increased. The greatest effects were seen with the highest of the 3 dose regimens (0.5–1 mg/kg/d). The smallest effect was observed for the intermittent dosing of 0.5 to 1 mg/kg/d for 1 week per month. Although the investigators suggested that changes in some of these hormone levels may be one of the mechanisms of action of isotretinoin in acne, some of these effects may also contribute to potential adverse effects, particularly because most treatment courses extend beyond 3 months. Further study of these hormonal changes is a priority.

Enhancing Bioavailability

Isotretinoin, a derivative of vitamin A, is highly lipophilic and bioavailability is enhanced with fat coingestion. Hence the standard practice recommendation of ingestion with food, particularly a high-fat meal, to enhance absorption. Initial pharmacokinetic studies of oral isotretinoin involved ingestion of a standardized high-fat meal comprising 1000 calories with 50% from fat. The bioavailability of standard oral isotretinoin formulations is approximately 60% lower in fasted conditions than after fatty meals, as described earlier.

It is inconceivable that such a meal will be taken on a daily basis by patients on oral isotretinoin in usual practice. Thus, recommendations in standard practice for coingestion of oral isotretinoin include foods with high fat content ( Table 1 ). Almost 1 in 3 adolescents do not have breakfast. Thus, inconsistent eating habits and inadequate fat content of coingestants may result in suboptimal and varying bioavailability.

A related issue is that the standard for bioequivalence of generic formulations is 80% to 125% of the innovator product Accutane (Roche Laboratories). Immediately before and for 2 decades after its approval for use in the United States in 1982, studies on dosing in acne clearance and potential for remission were done with the innovator, generic formulations being available in the United States only since 2002. If a cumulative threshold dose, as recommended by other investigators, is used as a therapeutic end point, there is a potential for underdosing while using generics.

Efficacy

The most rigorous studies regarding isotretinoin efficacy and adverse events in acne are from large randomized trials designed to minimize risk of bias. Within the constraints of that framework, only 2 studies over the past decade provide evidence on the potential of oral isotretinoin in achieving global success in acne.

In a noninferiority study of conventional oral isotretinoin versus oral isotretinoin-Lidose, the total cumulative dose attained was 126 mg/kg. Although the primary outcome was reduction in nodule counts on the face and trunk, information on rates of complete/almost complete clearance was also available. The proportions of subjects achieving complete/almost complete clearance with conventional oral isotretinoin and with isotretinoin-Lidose were 89% and 84%, respectively. A more recent study in severe nodular acne compared the relative efficacy and tolerability of oral doxycycline 200 mg daily plus fixed-dose adapalene 0.1%/benzoyl peroxide 2.5% gel with conventional oral isotretinoin. The resultant cumulative dose was also 126 mg/kg and the proportion of subjects attaining clear/almost clear facial acne in the oral isotretinoin arm was 77% (data on file, Galderma). For both studies, facial acne was not clear/almost clear in approximately 10% to 20% at end of treatment in the studies discussed earlier. It can be inferred that more prolonged treatment and correspondingly higher cumulative doses would be required to achieve complete clearance in all patients. Furthermore, these findings applied only to facial acne. Because truncal acne is known to lag in response to treatment compared to facial acne, it is anticipated that longer treatment durations, and therefore greater cumulative doses, are required for those with truncal involvement. Additional support that higher doses are likely required includes 2 studies in severe acne ( Table 2 ): 1 retrospective and another prospective in which the mean cumulative doses required were 290 mg/kg and 264 mg/kg, respectively.

Low and Alternative Dosing Regimens

The past decade has witnessed an increasing number of reports on oral isotretinoin in mild and moderate acne along with alternative dosing. Specifically these regimens have been characterized as minidose, comprising 20 mg up to 2 days per week; low dose, with daily dosing of 0.25 to 0.5 mg/kg/d; alternate day; or intermittent dosing, such as 7 or 10 days per month. The potential advantage of these regimens is maintenance of efficacy while reducing adverse events ( Table 3 ).

A recent study showed that improvement was similar for those receiving 0.5 to 0.7 mg/kg/d compared with 0.25 to 0.4 mg/kg/d and both were superior to 0.5 to 0.7 mg/kg/d once daily for 1 of every 4 weeks. Furthermore, relapses (defined as moderate or higher grade of acne) for the daily dosing groups were also similar and statistically superior to the intermittent dosing group. Adverse events were most frequently reported in the higher daily dose group, with cheilitis in 94% and xerosis in 31%, compared with the lower daily dose group, with rates of 65% and 6%, respectively. A shortcoming of that study is that an even higher, more conventional 0.7 to 1.0 mg/kg/d dose was not included.

Other studies have shown that lower doses in mild to moderate acne achieve excellent clearance of acne with high rates of remission despite attainment of cumulative doses less than 120 mg/kg. In an open-label prospective study to determine the efficacy of low-dose isotretinoin in inducing clearance and inducing remission of mild and moderate acne, dosing was initiated with 0.2 mg/kg/d and increased by 5 mg every 2 weeks to intolerance; intolerance was not further defined. Patients were treated for an additional month beyond complete acne clearance (mean cumulative dose of isotretinoin, 81 mg/kg) and maintained for 1 year on nightly adapalene 0.1% cream. At 2 years, relapse, defined as severity greater than mild, was observed in 13 of 139 patients (9%).

A recent study on low-grade adult acne showed the efficacy of isotretinoin 5 mg/d in reducing acne lesions and improving quality of life. Improvements in lesion counts were observed by 4 weeks and continued through the 32 weeks of treatment. At that juncture, 62.5% of subjects were completely clear of acne, defined as no acne lesions. Adverse events were primarily mild dryness of the skin and/or mucous membranes (62%) managed with topical moisturizers and lip balm.

Relapse

The initial recommendation of a minimum cumulative dose of 120 mg/kg for oral isotretinoin was based on observations of a reduced rate of relapse. The upper range of 150 mg/kg was established by a guidelines document indicating no further therapeutic gain with oral isotretinoin beyond that threshold. Subsequently, attainment of a cumulative dose of 120 to 150 mg/kg has been widely recommended as an end point for isotretinoin treatment with the goal of increasing the potential for remission. This axiom is present in the most contemporary of acne guidelines, oral isotretinoin review articles, and even in clinical research studies on new formulations of oral isotretinoin. Notably, it is absent in the most rigorously developed of current acne practice guidelines.

In a large database study linking medical service and pharmacy from the province of Quebec, Canada, predictors of acne relapse (defined as prescription of an acne medication) after oral isotretinoin were evaluated. The cohort comprised 17,351 first-time isotretinoin users with controls matched to each case. A total of 7100 (41%) subjects experienced an acne relapse (40.9%; 95% confidence interval [CI], 40.4–41.4), with half of this group receiving an acne medication within 59 months after completion of isotretinoin treatment. Of these, 26% required a second course of oral isotretinoin treatment. Male gender, age less than 16 years, urban domicile, isotretinoin cumulative doses less than 2450 mg (for a 70-kg patient, equivalent to 35 mg/kg), and treatment with isotretinoin less than 121 days were significantly associated with acne relapse. A progressive reduction in relapse rates with increasing cumulative doses of isotretinoin was noted such that the adjusted rate ratio of the highest dose range of greater than or equal to 7584 mg (for a 70-kg patient, equivalent to 108 mg/kg) was 0.63 (0.57–0.70) relative to those receiving less than 2450 mg. The follow-up duration of up to 20 years was longer than any prior study and likely reflects the true relapse rate. Although this study provided evidence of greater potential for remission with higher cumulative doses, information on baseline acne severity, daily dosimetry, and specific details on absolute relapse were unavailable. Furthermore, relapse, defined as prescription of acne medication, likely underestimates true acne relapse because some subjects may seek over-the-counter or alternative therapy, not present to their physicians for more treatment, or not fill their prescriptions for acne medication. Thus, an alternative perspective is that the 41% relapse rate may underestimate the true rate because there was inadequate information on acne severity, inadequate cumulative dosing to the recommendation of 120 to 150 mg/kg, and some may have had refractory acne wherein low-dose intermittent isotretinoin may have been prescribed.

In the observational study of 116 patients with severe nodulocystic acne, relapse rates at 1 year were lower in those receiving 220 mg/kg or more compared with those receiving lower doses (<220 mg/kg), with rates of 47.4% and 26.9%, respectively. Although a relapse rate of 33% was reported at 12-month follow-up, the article indicated that 65% of patients reported having had acne in the prior month. This discrepancy suggests that half the subjects with recurrent acne were not being treated with acne medications. Nonetheless, because physical examination was not performed on follow-up, the 65% recurrence rate reported by patients may be an underestimate. Thus, rate of recurrent acne in this study is likely within the range of 33% (use of prescription medication for acne) and 65% (patient self-report of acne).

A systematic review of the literature from 1980 to 2013 found a lack of high-quality studies supporting the cumulative dose of 120 to 150 mg/kg. On a 4-category scale of evidence quality from very low to high, the highest rated studies were 4 that were rated as moderate quality. Of these, none explicitly evaluated the cumulative dose of 120 to 150 mg/kg in acne relapse. Factors likely contributing to lack of high-quality evidence are inconsistencies in definition of key concepts, including acne clearance; treatment end point and relapse; as well as observational and retrospective study design. Nevertheless, when comparing multiple doses within the same study under controlled conditions, higher cumulative doses have consistently resulted in lesser relapse rates compared with lower doses.

Pregnancy Prevention

Recent investigations in women exposed to oral isotretinoin during pregnancy have provided new information on overall rates of exposure and risk for spontaneous abortions and malformations. In a study of 79 pregnant women referred for teratogen-risk counseling because of a history of isotretinoin exposure, 56 continued pregnancies to term. Of these, there were 11 spontaneous abortions, 44 healthy full-term babies, and 1 with multiple congenital abnormalities treated by therapeutic abortion. In the Netherlands, an exposure rate of 2.5 per 10,000 pregnancies was obtained from a cohort study of approximately 200,000 pregnancies over an 8-year period. In total there were 51 pregnancies exposed to isotretinoin: 45 during pregnancy and 6 within 30 days of discontinuing isotretinoin. An adverse outcome was observed in 5: 3 intrauterine deaths and 2 with major congenital anomalies.

The iPledge program is a computerized risk management system established in the United States in 2008 and requires 2 negative pregnancy tests before starting therapy and the use of 2 forms of contraception 30 days both before initiation of oral isotretinoin and on its cessation. Despite the iPledge program, between 2008 and 2011 there were still 150 women annually who became pregnant while on oral isotretinoin in the United States. A recent retrospective study found suboptimal concomitant rates of isotretinoin and contraceptive use before (29%) and after (32%) iPledge initiation. Adherence to contraception use continues to be an issue. An anonymous survey of 75 women of child-bearing potential who agreed to contraceptive methods or abstinence in iPledge found that of the 39 women who were sexually active, condom use had a noncompliance rate of 29%, whereas there was failure to use 2 forms of contraception in 44%. Of the 21 who chose abstinence, 7 had previously been sexually active, of whom 4 continued to be sexually active during treatment. Thus, even within a highly regulated pregnancy registry, there is evidence of ongoing risk behavior for pregnancy.

Recent studies have also examined contraceptive counseling methods by dermatologists. Interviews with 15 women participating in iPledge found that patients comprehended the adverse effect of isotretinoin on pregnancies and understood the risk of birth defects. Nevertheless, they received less information about effective means of contraception and none were fully informed about the advantages and disadvantages of all contraceptive options. Furthermore, many had misconceptions about highly effective reversible contraceptives such as intrauterine devices and subdermal implants. These methods of long-acting reversible contraception can lead to the same typical-use and perfect-use failure rates and are approved for up to 5 years of use.

Mental Health Disorders

The association of oral isotretinoin with depression and suicidal ideation has been addressed in multiple recent reviews. Based on the same body of research evidence, some investigators consider the potential for causality with oral isotretinoin in depression to be controversial, unsettled, potentially unresolvable, or strongly supportive. Comprehensive reviews of this issue in the literature focus on evidence from multiple perspectives, including temporality, outcomes of dechallenge-rechallenge, class effect of retinoids, dose responsivity, and biological plausibility. The evidence for biological plausibility comes from brain imaging studies and neurochemical pathways.

The literature on this association largely comprises case reports and retrospective studies, with few prospective trials. However, assessment of quality in these reviews is exclusively qualitative, none using quasiquantitative literature appraisal methodology. Of the prospective cohort studies, the 2 largest comprised 600 and 700 subjects with acne treated with oral isotretinoin. Both showed an identical depression rate of 1%. Of the retrospective studies, the largest was a case-control analysis involving 30,496 subjects. This study was the first to show a significant association between isotretinoin and depression, with a relative risk of 2.68 (CI, 1.10–6.48). Limitations of this study were absence of psychiatric confirmation of depression and inadequate determination of preexisting depression, particularly within the context of those with severe acne.

A subsequent review reported the incidence of major depression in adolescents as 3% to 11% and also highlights that acne is associated with depression or suicide. In the study of 3775 adolescents aged 18 to 19 years, 14% reported substantial acne (“a lot” or “very much” on a 4-category scale). Compared with subjects with no/little acne, suicidal ideation was 2-fold and 3-fold higher in girls and boys, respectively. Mental health problems, including depression, were also associated with substantial acne. Potential confounding with oral isotretinoin was mitigated by the low rate of prescribing in this cohort. The investigators concluded that acne, frequently found in late adolescence, is associated with social and psychological problems. Furthermore, suicidal ideation and depression, associated with acne therapy such as oral isotretinoin, may reflect the burden of substantial acne rather than an effect of medication.

The effect of severe acne as distinct from that of oral isotretinoin was also addressed by a study of 5756 patients with severe acne. Time frames were divided into junctures before, during, and after treatment with isotretinoin. The risk of suicide attempts was found to progressively escalate from 3 years before isotretinoin treatment and remained increased until 6 months posttreatment. Thus severe acne, independent of oral isotretinoin, contributes to an increased risk of depression and suicide.

All patients should be informed about uncommon risk for mood and behavioral changes associated with acne and possibly with oral isotretinoin. A brief practical and validated screening instrument with high sensitivity and specificity for major depressive disorders can be of value in practice, such as the 2-item Patient Health Questionnaire (PHQ-2; Copyright Pfizer Inc). In patients with positive screen findings, further evaluation with more specific instruments, such as the 9-item PHQ-9 (Copyright Pfizer Inc) or Hamilton Depression Rating Scale can confirm the diagnosis and provide determination of severity. Further management with the family physician and/or a mental health professional is recommended to develop a strategy for ongoing monitoring and mitigation of mental health risk if oral isotretinoin is prescribed. For these patients, consider initiation with a low starting dose of oral isotretinoin and slow escalation.

Inflammatory Bowel Disease

Before 2013, there were 4 epidemiologic studies and 2 abstracts evaluating the association between oral isotretinoin and inflammatory bowel disease (IBD). Two that were initially performed to address case reports of association provided divergent results: one showing no association, the other showing strong association with ulcerative colitis (UC) but not Crohn disease (CD). However, findings from the latter could not be replicated by a subsequent study controlling for prior oral antibiotic use and acne severity while reducing confounding by oral contraceptive use. Another large epidemiologic study from British Columbia, Canada, found no association between oral isotretinoin and IBD. However, a weak association was detected for both topical acne medications and oral isotretinoin with IBD in young subjects (aged 12–19 years), suggesting that the association may be with acne itself rather than its treatment.

Three recent studies also do not show an association of IBD with oral isotretinoin exposure. In a study evaluating risk of IBD in women of reproductive age with use of isotretinoin using a health claims database, the adjusted relative risk for UC was 1.10 (95% CI, 0.44–2.70) and for CD was 0.91 (95% CI, 0.37–2.25). A French National Health Insurance case-control study involving 7593 cases of IBD found no association between oral isotretinoin and UC. Odds ratio (OR) for UC was 1.36 (95% CI, 0.76–2.45) and for CD was 0.45 (95% CI, 0.24–0.85). In a retrospective study of patients with acne seeking treatment over a 5-year period, the risk of IBD was lower in the isotretinoin-exposed group (unadjusted OR, 0.33; 95% CI, 0.12–0.93; P = .04).

Oral antibiotics are associated with IBD based on the results of a meta-analysis of 11 observational studies involving 7208 patients. The overall ORs for antibiotic exposure and CD and UC were 1.74 (95% CI, 1.35–2.23) and 1.08 (95% CI, 0.91–1.27), respectively. In children, an OR for CD was observed of 2.75 (95% CI, 1.72–4.38). All antibiotics, except penicillin, were associated with IBD and highest risks were for metronidazole (OR, 5.01; 95% CI, 1.65–15.25) and fluoroquinolones (OR, 1.79; 95% CI, 1.03–3.12).

Abnormal Wound Healing

Current recommendations to avoid acne scar repair procedures within 6 months after oral isotretinoin arose from reports in the mid-1980s of delayed wound healing and hypertrophic scarring with conventional and argon laser dermabrasion. However, recent studies of resurfacing procedures suggest that wound healing is not impaired if treatment is performed earlier. In 10 patients treated within 1 to 3 months after oral isotretinoin with medium-depth peels and manual sandpaper dermabrasion, 6-month follow-up showed no evidence of hypertrophic or keloidal scars. In 20 patients treated with ablative carbon dioxide laser within 1 to 3 months after oral isotretinoin, no hypertrophic scars/keloids were noted at 6-month follow-up. An additional study of patients on low-dose isotretinoin (10 mg/d) showed no abnormal scars after nonablative infrared fractional laser.

Recent animal studies have corroborated these findings. In rabbits, 40 wounds created with 6-mm punch biopsies in the ears of 2 control and 2 experimental rabbits (the latter fed isotretinoin 4 mg/kg/d) showed no difference in wound healing clinically or histologically. Collagen synthesis was also similarly unaffected. In minipigs, a model with similarity to human cutaneous structure and wound repair, standard full-thickness (punch biopsies) and partial-thickness (phenol peel) injuries were evaluated over 28 days in an animal treated with isotretinoin (2 mg/kg/d over a 60-day period, attaining a cumulative dose of 120 mg/kg) and a control without treatment. Histologic and photographic measures of wound healing did not differ at 7, 14, or 28 days.