Discussion
To our knowledge, this is the first prospective evaluation of MA after NSAI failure in endocrine-responsive breast cancer. MA is an active treatment in this setting, with a CBR of 40%, a median duration of clinical benefit of 10.0 months, and a median PFS of 3.9 months.
Our results compare favorably with other trials that evaluated single hormonal agents after breast cancer progression on a NSAI ( Table 3 ) . As resistance to endocrine therapy may be related to the activation of other signaling pathways, Baselga et al. evaluated the combination of exemestane with everolimus, a sirolimus derivative that inhibits mTOR, in a randomized phase III trial. The addition of everolimus improved PFS to 6.9 months, when compared with 2.8 months in the placebo arm (hazard ratio 0.43; 95% CI 0.35–0.54; P < 0.001). At the same time, the two-drug combination was associated with a higher incidence of adverse events: 19% of patients discontinued everolimus due to toxicity and 5% withdrew consent.
PFS in the current study did not vary according to the presence or absence of visceral metastases, although this analysis was underpowered. Randomized clinical trials of single endocrine agents are in agreement with our results and the presence of visceral metastases should not preclude the use of endocrine therapy .
The androgen receptor is frequently co-expressed with the ER and its expression is maintained in the metastastic setting. There is evidence suggesting the AR to be prognostic as well as predictive of progestin treatment effect. We did not show a correlation between expression of the androgen receptor and MA efficacy, although the small number of specimens available may have compromised the analysis.
The treatment benefits from MA occurred at the expense of moderate toxicity. There were no grade 4 adverse events and few patients presented with grade 3 toxicity. The most relevant untoward side-effects were weight gain and DVT. We observed only one patient with weight gain >20%, but >40% of patients had at least 10% increase in weight above baseline. Recent comparisons of MA with NSAI in advanced breast cancer showed average weight gains in the range of 9%–17%. Weight gain may become more pronounced over time, which represents an inconvenience for patients who continue treatment of longer periods.
Thromboembolic disease is another concern related to MA use. When compared with NSAI in advanced breast cancer, the incidence of combined superficial thrombophlebitis, DVT, and pulmonary embolism varied from 5% to 8%. We observed a DVT rate of 10% in this patient population. The higher rate documented in our trial may be related to the prospectively scheduled assessment through Doppler ultrasound of the lower extremities leading to the diagnosis of asymptomatic events. All patients who developed a DVT had clinical benefit on MA, with median treatment duration of 11 months. Therefore, we encourage a low threshold for the evaluation of DVT for the patients who remain on treatment.
Our study main limitation includes the unknown hormone receptor status for 17% of the patients (due to transfer from other hospitals and no remaining tissue availability) as well as receptor evaluation done at diagnosis and not at the time of disease progression. A switch in ER status, in either direction, occurs in ~10% between primary tumor and recurrence. However, clinically defined hormone sensitivity as entry criteria may have circumvented these shortcomings.
ER-positive breast cancer is a heterogeneous group of diseases with considerable variability in outcome to a range of treatments. Prior response predicts the likelihood of subsequent benefit from another endocrine agent, and this should be taken into account in the treatment decision process when assessing whether to prescribe a subsequent endocrine therapy. Some patients have extended hormone-sensitive disease that benefits from multiple endocrine regimens administered sequentially. We showed that MA lacks cross-resistance and is active after acquired resistance to potent AI. However, as there are no data showing that one drug sequence is preferable to another, no definitive recommendations can be made. At the same time, despite the benefits from new pharmacologic treatments, many of them may be unavailable or limited by cost constraints. It is well recognized that treatment choices have to be made within financial boundaries and physicians may consider selecting a cheaper drug in a cost-sensitive environment. As such, MA, an off-patent drug, should be entertained in the treatment algorithm as a valuable option.
In summary, MA has demonstrated activity with moderate toxicity as the treatment of postmenopausal women after progression on a NSAI. Therefore, MA should be considered as an alternative in the sequence of agents for hormone-sensitive breast cancer. Despite the lack of benefit from recent reports on progesterone receptor antagonists, our results support the rationale for continuing to pursue progestins as a fruitful strategy in breast cancer treatment.