Furthermore, when the data were analyzed according to specific HMT agent, patients receiving tamoxifen or an aromatase inhibitor had significant reductions in risk of Alzheimer disease compared with those not receiving HMT, whereas those receiving raloxifene did not.
From the paper, an overview of what HMT does:
Hormone-modulating therapies (HMTs) include the selective estrogen receptor modulators (SERMs; tamoxifen and raloxifene) and aromatase inhibitors (steroidal, exemestane; nonsteroidal, anastrozole and letrozole). These drugs have been used for the treatment of estrogen receptor–positive breast cancers and have been shown to decrease estrogen’s effects at the level of the breast tissue.13 Tamoxifen is used in both the treatment and the prevention of estrogen receptor–positive breast cancer and is a common therapy for premenopausal women and an option for postmenopausal women.12
They found more specifically:
For patients 65 to 69 years of age, there was no significant difference between patients receiving HMT and patients not receiving HMT in the risk of NDD or AD; SDs overlapped in the 5-year analysis. In contrast, increasing age was associated with a greater reduction of risk for all NDDs in women receiving HMT; SDs did not overlap in the 5-year analysis and were divergent.
But here's their theory - while decreasing estrogen in breast cells, tamoxifen and aromatase inhibitors have divergent effects in the brain, ultimately increasing estrogen.
If tamoxifen and aromatase inhibitors are acting to increase estrogen-related actions in brain tissue, the argument for the protective association of estrogen with AD-related outcomes is strengthened.
Full paper here.