To the Editor:
Gynecomastia is an emerging condition in HIV-infected patients and could be related to lipodystrophy and dysmetabolism, especially when protease inhibitors (PIs) are used. 1,2 When anti-HIV drugs other than PIs became available, regimens sparing or excluding PIs and based on triple nucleoside analogues (NAs), especially nonnucleoside reverse transcriptase inhibitors, became increasingly popular. 3–6 Some PI-sparing combinations were assessed just in patients who developed dysmetabolism and/or lipodystrophy during PI treatment, but most results were contradictory, probably because of reduced patient samples (16–104/cases) and limited follow-up (6–12 months). A switch from PI-containing highly active antiretroviral therapy (HAART) to nonnucleoside reverse transcriptase inhibitor– or abacavir-containing HAART was effective for dysmetabolism but was infrequently effective for lipodystrophy. 1 After the first case of gynecomastia during NA use was reported, 7 other episodes occurred in the absence of PI treatment, 2,8 and the appearance or worsening of dysmetabolism and/or lipodystrophy after discontinuation of PI therapy was increasingly reported, 1,3,9 especially in association with efavirenz treatment. 10
A cross-sectional survey of 981 patients treated with antiretrovirals for ≥12 months (658 males [67.1%]) was performed in August 2003 to identify all subjects with true (ultrasonography-confirmed) gynecomastia. Patients with adherence levels of <90% (as reported by patients, specific questionnaires, and direct drug distribution; accountability was carried out monthly) were excluded, as were subjects with chronic liver and/or kidney disease, subjects with endocrine disorders, and subjects using drugs that may induce gynecomastia (including marijuana). The eventual lipodystrophy syndrome was evaluated by anthropometric measurement, patient questionnaire, bioelectric impedance assay, and DEXA. The endocrinologic workup included follicle-stimulating hormone, luteinizing hormone, prolactin, testosterone, estradiol, and dehydroepiandrosterone assays.
Since 1999, 14 (2.85%) of 492 evaluable males developed gynecomastia at the ages of 12–58 years; HIV infection was sexually acquired in all cases but 1 and was known since 19–183 months (Table 1). This disorder was perceived with esthetic and psychologic discomfort, was accompanied by local hyperesthesia and/or mild pain in 9 patients, and was predominantly bilateral in 10 patients. In all patients, gynecomastia concurred with the fat redistribution syndrome, characterized by moderate-to-severe peripheral fat wasting, thinning of the thighs, and increased abdominal girth; predominant peripheral lipoatrophy was present in patient 6, who received prolonged isolated dual NA therapy. Abnormal serum triglyceride levels occurred in 10 cases; hypercholesterolemia, in 6; and hyperglycemia, in 3: on the whole, only 2 of 14 patients with gynecomastia did not have dysmetabolism (patients 2 and 14). When gynecomastia was recognized, antiretroviral therapy had been taken from 23 to 142 months and was initiated with NAs only in 11 cases. The administration of PI-based regimens lasted for a mean ± SD of 30.7 ± 23.2 months (range, 6–83 months) before the occurrence of gynecomastia in only 9 patients (64.3%): 7 received >1 PI-based regimen, and 2 developed gynecomastia during their first PI-containing regimen (lopinavir/ritonavir, patient 2; indinavir, patient 3). As a result, 5 (35.7%) of 14 gynecomastia episodes occurred in patients who never received PIs: 4 involved patients whose first HAART regimen included the nonnucleoside reverse transcriptase inhibitor efavirenz (patients. 1, 8, 11, and 12), and 1 involved a patient (patient 6) who underwent a 62-month regimen with dual NAs only (lamivudine and stavudine taken since 39 months; gynecomastia developed 17 months after the introduction of this regimen) (Table 1). On the whole, 12 (85.7%) of 14 patients were receiving stavudine at the time of diagnosis of gynecomastia, with a treatment duration of 9–39 months (mean ± SD, 16.1 ± 6.3 months). The 2 remaining patients were treated with NAs other than stavudine but received stavudine until 4–9 months before the occurrence of gynecomastia for a 12- to 19-month period. No other NA was used so extensively by our patients with gynecomastia.
Laboratory markers of HIV disease showed complete viral suppression in 11 cases and residual viremia (HIV RNA level, 680–4200 copies/mL) in 3, while CD4+ lymphocyte counts ranged from 421 to 778/μL. A prior or concurrent diagnosis of AIDS was excluded for all patients but 1 (patient 5) (Table 1). No significant correlation between gynecomastia and virologic, immunologic, and clinical markers of HIV disease was found. Endocrine abnormalities were excluded, except for isolated mild increases in follicle-stimulating hormone and luteinizing hormone levels in patient 5; serum prolactin levels were always normal. The subsequent mean follow-up ± SD of 9.3 ± 4.2 months (range, 8–28 months) was characterized by a stable course of gynecomastia in 11 cases (78.6%), even after modification of anti-HIV therapy, and a mild-to-moderate worsening and/or bilateral extension in patients 2, 3, and 7, who switched from PI- to efavirenz-based HAART. The 12-year-old child (patient 13) developed bilateral gynecomastia after 7 years of PI treatment, with the last 31-month regimen containing stavudine, didanosine, and nelfinavir. Gynecomastia was not so severe that HAART had to be discontinued or surgery was required. Paralleling persisting gynecomastia, lipodystrophy and dysmetabolic alterations did not significantly improve after HAART changes except for a reduction in serum lipid levels prompted by diet, physical exercise, and eventual hypolipidemic drugs.
Over 50 cases of gynecomastia have been reported in association with HIV infection and antiretroviral therapy, 2,7,8,11–13 although some reports did not exclude all pharmacologic risk factors. 8,13 Most cases of HIV-associated gynecomastia were bilateral, but monolateral episodes predominantly occurred. 8,13 The differential diagnosis of HIV-associated gynecomastia should rule out “lipomastia” (pseudogynecomastia) and rare malignancies (breast lymphoma may resemble gynecomastia). 13 Most reports indicated prior PI administration, while rare series emphasized a possible role of prolonged NA use. 2,7 When evaluating the composition of HAART regimens that eventually prompted gynecomastia, the NA stavudine seemed to be the most frequently used drug, 2,7,8 as confirmed in our series. In the first reported case apparently associated with stavudine, 7 gynecomastia resolved just after drug discontinuation. Anecdotal reports of efavirenz-associated gynecomastia have been emerging in the absence of lipodystrophy, 10–12 such that novel pathogenetic mechanisms were postulated for this untoward effect (including the immune reconstitution syndrome). 14 In our experience, efavirenz introduction apparently prompted gynecomastia in 4 PI-naive patients and in 3 more subjects who switched from PIs to efavirenz-based regimens. We emphasize the emergence of gynecomastia in patients who never received (or discontinued) PI treatment as well as the frequent persistence, appearance, or worsening of gynecomastia during efavirenz therapy both in HAART-naive patients and in patients switching from PIs. To our knowledge, we also report the first case of pediatric gynecomastia.
Compared with the prevalence observed by Piroth et al 15 (0.8% of HAART-treated males, with increasing figures according to treatment duration), we observed a significantly greater crude frequency (2.85%). Gynecomastia usually persisted despite therapeutic changes during a prolonged mean follow-up (9.3 months), while the association with fat redistribution syndrome and/or hyperlipidemia was predominant, compared with previously reported findings. 8,11,12 The lack of recovery, fixed association with lipodystrophy, and high frequency of concurrent dysmetabolism may lead one to suppose common pathogenetic mechanisms. On the other hand, no correlation was found between the occurrence and features of gynecomastia and virologic, immunologic, and clinical markers of HIV disease. A 21.4% rate of hyperprolactinemia was detected among 192 HIV-infected men, although this occurrence was not related to HAART, liver disease, metabolic disturbances, viremia, and gynecomastia 16; in our patients, no prolactin alteration was found. The hormone profile was normal for almost all evaluated patients, as previously reported 7,11,12; therefore, the pathogenesis of gynecomastia remains unclear, with a large spectrum of hypotheses, including decreased estrogen metabolism or testosterone synthesis, changes in the estrogen/androgen ratio, direct drug effect on the breast tissue, subtle metabolic/endocrine abnormalities, or altered cytokine network 7,11,12 as well as immune reconstitution. 14 Although most studies focused on the relationship between gynecomastia and PIs, 2,8,13,15 the extended number of HAART combinations and the increasing descriptions of gynecomastia in patients who never received (or discontinued) PIs make the assessment of epidemiology, temporal evolution, and, especially, pathogenetic mechanisms underlying this untoward event intriguing. 1,3,9–12,15
Gynecomastia, as well as lipodystrophy and metabolic abnormalities, may occur regardless of the extensive use of PIs. The eventual pathogenetic pathways shared among gynecomastia, female breast hypertrophy, lipodystrophy, and dysmetabolism deserve further investigation, especially after the occurrence or worsening of gynecomastia associated with PI-sparing regimens and the persistence of metabolic or fat redistribution alterations even after discontinuation of PI treatment. Switching from a PI-based regimen to those containing efavirenz may not guarantee significant benefits for gynecomastia, lipodystrophy, and dyslipidemia, because in our experience 50% of patients developed (or had worsening) of these disturbances after initiation of efavirenz therapy. Therefore, the role of efavirenz in this condition warrants investigation, because this drug alone is strongly recommended as a potent and safe alternative for patients discontinuing PI-based HAART because of failure or toxicity 1,3–6 and differences with other nonnucleoside reverse transcriptase inhibitors are still under investigation. 1,9 The pathogenetic mechanisms of lipodystrophy, dyslipidemia, and gynecomastia arising or persisting during nonnucleoside reverse transcriptase inhibitor treatment are unknown, especially for differences among single compounds. Controlled studies are needed to assess both the frequency and the pathogenesis of gynecomastia and related untoward events during different regimens and to reliably evaluate the emerging problem of gynecomastia, which poses relevant cosmetic and psychologic problems with expected consequences on patient adherence to anti-HIV regimens. We emphasize the appreciable frequency of gynecomastia, its occurrence in the absence of PI therapy, its persistence despite therapeutic changes (thus resembling some features of the lipodystrophy syndrome), the apparent association with prolonged NA administration (especially stavudine), and the possible correlation with efavirenz.
Roberto Manfredi, MD
Leonardo Calza, MD
Francesco Chiodo, MD
1. Martínez E, Conget I, Lozano L, et al. Reversion of metabolic abnormalities after switching from HIV-1 protease inhibitors to nevirapine. AIDS. 1999; 13:805–810.
2. Manfredi R, Calza L, Chiodo F. Gynecomastia associated with highly active antiretroviral therapy. Ann Pharmacother. 2001; 35:438–439.
3. Negredo E, Cruz L, Paredes R, et al. Virological, immunological, and clinical impact of switching from protease inhibitors to nevirapine or to efavirenz in patients with human imunodeficiency virus infection and long-lasting viral suppression. Clin Infect Dis. 2002; 34:504–510.
4. Chiesa E, Bini T, Adorni F, et al. Simplification of protease inhibitor-containing regimens with efavirenz, nevirapine or abacavir: safety and efficacy outcomes. Antivir Ther. 2003; 8:27–35.
5. Asboe D, Mandalia S, Gazzard BG. Sequencing to NRTI plus NNRTI-only combinations after virological failure of protease inhibitor-based combination HIV-1 therapy. HIV Clin Trials. 2003; 4:1–10.
6. Perez-Molina JA. Safety and tolerance of efavirenz in different antiretroviral regimens: results from a national multicenter prospective study in 1,033 HIV-infected patients. HIV Clin Trials. 2002; 3:279–286.
7. Melbourne KM, Brown SL, Silverblatt FJ. Gynecomastia with stavudine treatment in an HIV-positive patient. Ann Pharmacother. 1998; 32:1108.
8. Paech V, Lorenzon T, Von Krosigk A, et al. Gynaecomastia in HIV-infected men: association with effects of antiretroviral therapy. AIDS. 2002; 16:1193–1195.
9. Ruiz L, Negredo E, Domingo P, et al. Antiretroviral treatment simplification with nevirapine in PI-experienced patients with HIV-associated lipodystrophy: one-year prospective follow-up of a multicentre, randomized, controlled study. J Acquir Immune Defic Syndr. 2001; 27:229–236.
10. Rickerts V, Bickel M, Tews J, et al. A case of lipodystrophy after the initiation of zidovudine, lamivudine and efavirenz in an asymptomatic HIV-1-infected patient. HIV Med. 2000; 1:164–165.
11. Caso JA, Prieto Jde M, Casas E, et al. Gynecomastia without lipodystrophy syndrome in HIV-infected men treated with efavirenz. AIDS. 2001; 15:1447–1448.
12. Qazi NA, Morlese JF, King DM, et al. Gynaecomastia without lipodystrophy in HIV-1-seropositive patients on efavirenz: an alternative hypothesis. AIDS. 2002; 16:506–507.
13. Evans DL, Pantanowitz L, Dezube BJ, et al. Breast enlargement in 13 men who were seropositive for human immunodeficiency virus. Clin Infect Dis. 2002; 35:1113–1119.
14. Qazi NA, Morlese JF, King DM, et al. True gynaecomastia, another manifestation of immune reconstitution disease? Int J STD AIDS. 2002; 13:59–65.
15. Piroth L, Grappin M, Petit JM, et al. Incidence of gynecomastia in men infected with HIV and treated with highly active antiretroviral therapy. Scand J Infect Dis. 2001; 33:559–560.
16. Collazos J, Ibarra S, Martínez E, et al. Serum prolactin concentrations in patients infected with human immunodeficiency virus. HIV Clin Trials. 2002; 3:133–138.