You should try tamoxifen.
http://www.endotext.org/male/male14/male14.htm
MEDICAL TREATMENT
If the gynecomastia is severe, does not resolve, and does not have a treatable underlying cause, some medical therapies may be attempted. These include testosterone, dihydrotestosterone, danazol, clomiphene citrate, tamoxifen and the aromatase inhibitor testolactone. Testosterone treatment of hypogonadal men with gynecomastia often fails to produce breast regression once gynecomastia is established. Unfortunately, testosterone treatment may actually produce the side effect of gynecomastia by being aromatized to estradiol. Thus, although testosterone is used to treat hypogonadism, its use to specifically counteract gynecomastia is limited (42). Dihydrotestosterone, a non-aromatizable androgen, has been used in patients with prolonged pubertal gynecomastia with good response rates (22). Since dihydrotestosterone is given either intramuscularly or percutaneously, this may restrict its usefulness. Danazol, a weak androgen that inhibits gonadotropin secretion, resulting in decreased serum testosterone levels, has been studied in a prospective placebo-controlled trial, whereby gynecomastia resolved in 23 percent of the patients, as opposed to 12 percent of the patients on placebo (20). Unfortunately, undesirable side effects including edema, acne, and cramps have limited its use (27). Investigators have reported a 64 percent response rate with 100 mg/day of clomiphene citrate, a weak estrogen and moderate antiestrogen (24). Lower doses of clomiphene have shown varied results, indicating that higher doses may need to be administered, if clomiphene is to be attempted. Tamoxifen, also an antiestrogen, has been studied in 2 randomized, double-blind studies in which a statistically significant regression in breast size was achieved, although complete regression was not documented (1). One study compared tamoxifen with danazol in the treatment of gynecomastia. Although patients taking tamoxifen had a greater response with complete resolution in 78 percent of patients treated with tamoxifen, as compared to only a 40 percent response in the danazol-treated group, the relapse rate was higher for the tamoxifen group (41). Although complete breast regression may not be achieved and a chance of recurrence exists with therapy, tamoxifen, due to relatively lower side effect profile, may be a more reasonable choice when compared to the other therapies. If used, tamoxifen should be given at a dose of 10 mg twice a day for at least 3 months (27). An aromatase inhibitor, testolactone, has also been studied in an uncontrolled trial with promising effects (45). Further studies must be performed on this drug before any recommendations can be established on its usefulness in the treatment of gynecomastia.
----------------------
http://bmj.bmjjournals.com/cgi/content/full/327/7410/301
The use of tamoxifen for gynaecomastia has been studied previously in several centres. The table shows the various published studies on the use and efficacy of tamoxifen for physiological gynaecomastia in the English literature.6-9 Only two of these studies6 9 have more than 10 patients and both showed resolution of lump and pain in 80% of cases. A recent study from our own unit in 36 cases confirms this figure (83% resolution of lump).10 Ting et al also found tamoxifen to be more efficacious than danazol.6 Importantly only minor and reversible side effects were reported. This confirms findings that tamoxifen used in male breast cancer appears to have no serious side effects.11
Tamoxifen appears to be successful, safe, and avoids operation and on present evidence should be regarded as the first line treatment of gynecomastia.