Prostate cancer is one of the most common cancers among men in China. The morbidity and mortality rates of prostate cancer were reported to be 2.74 and 2.26% in 2013; however, these rates have decreased over the last 10 years due to prostate-specific antigen (PSA) screening. The standard first-line treatment of metastatic prostate cancer is androgen-deprivation therapy (ADT) (1). However, the majority of the patients develop progressive disease in 14–30 months, which is resistant to ADT and referred to as castration-resistant prostate cancer (CRPC).

The optimal therapy for patients with CRPC is considered to be chemotherapy with taxane-based regimens, abiraterone and sipuleucel-T (2–4). However, patients aged ≥70 years may not be suitable for these therapies and treatment selection should be carefully considered due to the different patient status. A phase II clinical study demonstrated that the combination of granulocyte-macrophage colony-stimulating factor (GM-CSF) and thalidomide may decrease the level of PSA and exert antitumor effects (5).

To the best of our knowledge, our study was the first to evaluate the efficacy and determine the mechanism of action of GM-CSF plus thalidomide in the treatment of CRPC patients aged ≥70 years in China.

The recommended therapy for patients with CRPC is chemotherapy with taxane-based regimens, abiraterone and sipuleucel-T (2–4). However, elderly patients may not be suitable for these therapies and treatment should be carefully selected. A phase II clinical study demonstrated that the combination of GM-CSF and thalidomide effectively decreased the level of PSA and exhibited antitumor activity (5).

The antitumor effect of thalidomide is attributed to the inhibition of tumor angiogenic activity, induction of apoptosis in vitro and reduction of the high levels of angiogenic factors, such as vascular endothelial growth factor and basic fibroblast growth factor in patients with prostate cancer (6–8). Thalidomide was primarily assessed by a decline in PSA expression at doses of 100–200 mg/day (8). A phase 2 study reported that low-dose thalidomide (100 mg daily for ≤6 months) was effective in reducing PSA levels: 37.5% of the treated patients exhibited a median decrease in PSA levels of 48%, whereas a decrease of 50% was sustained throughout treatment for 15% of the patients. Another phase 2 trial compared low-dose (200 mg/day) and high-dose (≤1,200 mg/day) thalidomide in patients with metastatic androgen-independent prostate cancer (AIPC). Serum PSA decreases of 50% were observed in 18% of the patients in the low-dose arm. Patients in the high-dose arm exhibited no PSA decrease, although the side effects limited dose escalation to >200 mg/day in 30% of the cases. A total of 27% of the patients exhibited a PSA decrease of 40%, often associated with improvement of the clinical symptoms (8,9).

Prostate cancer cells kill mature dendritic cells (DCs), inhibit DC proliferation, differentiation and maturation, thereby reducing the number of DCs and inhibiting their function. DC deficiency and dysfunction is a major cause of prostate cancer immune evasion. Small et al (10) demonstrated the biological activity of GM-CSF (PSA decrement) in a series of androgen-independent and hormonal therapy-naive prostate cancer patients. Rini et al (11,12) investigated the biological effect of GM-CSF (250 µg/m²/day), as measured by PSA kinetics, in 30 patients with increasing PSA levels following radical prostatectomy or radiotherapy. Of the 29 evaluable patients, 3 (10%) achieved a 50% reduction in PSA [95% confidence interval (CI): 2–27%]. In the 26 patients whose pre-treatment PSA doubling time (PSADT) was calculated, the median PSADT increased from 8.4 to 15 months (P﹤0.001) and the median slope of the PSA vs. time curve decreased with treatment (P﹤0.004). Of the 29 evaluable patients, 7 (24%) remained free of disease progression after a median of 4.4 years (range, 4.0–4.8 years) of GM-CSF treatment.

Dreicer et al (5) conducted a phase 2 trial of thalidomide plus GM-CSF in 22 patients with metastatic AIPC. GM-CSF (250 µg) was administered 3 times per week, with thalidomide titrated to 200 mg/day, for 6 months. All 22 patients exhibited decreased PSA levels 2 weeks after treatment. A total of 5 patients exhibited a 50% decrease in PSA from baseline (median, 58 ng/ml), verified at 4 weeks after best response. The observed response rate was 23% (95% CI: 8–45%).

Amato et al (13) performed another phase 2 trial of thalidomide plus GM-CSF in 21 patients with hormone-naïve prostate cancer. GM-CSF (250 µg/m²; maximum, 500 µg) was administered 3 times per week, with thalidomide titrated to 100 mg/day (maximum, 400 mg). In the 18 patients who responded to this treatment, the median PSA reduction was 59% (range, 26–89%) and 12 patients (67%) exhibited a reduction of ≥50%.

In our study, the median age of the 11 patients was 77 years and, as determined by the NCCN guidelines, the patients were not suitable for chemotherapy with taxane-based regimens. In addition, abiraterone and sipuleucel-T have not been approved for use in China. After obtaining permission from the patients and their families, GM-CSF plus thalidomide was selected as the treatment regimen. The results indicated that this combination was clinically effective in older (aged ≥70 years, with a life expectancy of >1 year) patients with CRPC (Fig. 1). In our study, 3 patients (27.2%) had a PSA rebound, with 1 patient exhibiting a PSA rebound of >50% (Table III) in cycle 2; 1 patient (27.2%) had a PSA rebound of <50% (Table IV) in cycle 3; and 2 patients (18.2%) had PSA rebounds of >50% (Table V) in cycle 4. PSA rebound was attributed to extensive tumor cell necrosis caused by the treatment, which may exhibit significant individual differences; therefore, the changing trend of PSA during treatment should be the main focus rather than the PSA level. In responders, the overall median decrease in PSA was 92.2%, although several patients exhibited a PSA rebound in different cycles. The regimen was well tolerated, with mild (grade 1–2) toxicities. The most frequent grade 1–2 toxicities were fever, nausea, constipation, neuropathy and fatigue, whereas 1 patient developed grade 1 deep venous thrombosis.

In conclusion, our study demonstrated that the combination of GM-CSF plus thalidomide was clinically effective and well tolerated and may be considered an optimal treatment choice for elderly patients with CRPC.