A number of men have asked or commented about Keytruda, including Gus, Blair77 & DSim.
New study below.
The target of Merck's Keytruda (Pembrolizumab) is the receptor known as PD-1 (Programmed cell death protein 1), which suppresses T cell activity, thereby dampening the immune response.
The concept behind Keytruda as a viable product is that, while only a minority with a specific cancer type will benefit, the potential market is quite large. Thus, it is a general cancer treatment, that requires pre-screening.
Baas, et al, 2017 , looked for PD-1, PD-L1 (its natural ligand), and CD3 (T cell marker) in "tissue samples taken from 25 men with high-grade prostate cancer."
"An overall low expression of PD-1 and PD-L1, and a concurrent high expression of CD3+ T cells was found in high-risk prostate cancer tissue." "Because of this, one might be able to question the role of PD-L1 in local immune suppression in prostate cancer."
From a 2016 paper :
"While programmed cell death 1 (PD-1) inhibitors have shown clear anti-tumor efficacy in several solid tumors, prior results in men with metastatic castration resistant prostate cancer (mCRPC) showed no evidence of activity. Here we report unexpected antitumor activity seen in mCRPC patients treated with the anti-PD-1 antibody pembrolizumab. Patients with evidence of progression on enzalutamide were treated with pembrolizumab 200 mg IV every 3 weeks for 4 doses; pembrolizumab was added to standard dose enzalutamide. Three of the first ten patients enrolled in this ongoing phase II trial experienced rapid prostate specific antigen (PSA) reductions to ≤ 0.2 ng/ml."
A 2015 Canadian paper  had found PD-L1 to be highly expressed in Enzalutamide resistant prostate cancer.
It appears that PD-1 activation might be a common resistance response to Xtandi. So Keytruda, in essence, would simply be extending the effectiveness of Xtandi? Or is there added value?
In the new study:
"The genomes of cancers deficient in mismatch repair (MMR) contain exceptionally high numbers of somatic mutations."
We evaluated the "efficacy of PD-1 blockade in patients with advanced MMR-deficient cancers across 12 different tumor types. Objective radiographic responses were observed in 53% of patients and complete responses were achieved in 21% of patients."
From a NY Times article (today) :
"After taking pembrolizumab, 66 patients had their tumors shrink substantially and stabilize, instead of continuing to grow. Among them were 18 patients whose tumors vanished and have not returned."
"The drug, made by Merck, is already on the market for select patients with a few types of advanced lung, melanoma and bladder tumors. It is expensive, costing $156,000 a year.
A test for the mutations targeted by the drug is already available, too, for $300 to $600.
Just 4 percent of cancer patients have the type of genetic aberration susceptible to pembrolizumab."
At the same time:
"Merck $MRK has had to hit the brakes on enrolling new patients for two of its late-stage combo studies involving its blockbuster checkpoint Keytruda after the monitoring committee raised a red flag on an imbalance of deaths in the studies involving multiple myeloma." 
The genomes of cancers deficient in mismatch repair (MMR) contain exceptionally high numbers of somatic mutations. In a proof-of-concept study, we previously showed that colorectal cancers with MMR deficiency were sensitive to immune checkpoint blockade with anti-PD-1 antibodies. We have expanded this study to now evaluate efficacy of PD-1 blockade in patients with advanced MMR-deficient cancers across 12 different tumor types. Objective radiographic responses were observed in 53% of patients and complete responses were achieved in 21% of patients. Responses were durable with median progression-free and overall survival still not reached. Functional analysis in a responding patient demonstrated rapid in vivo expansion of neoantigen-specific T cell clones that were reactive to mutant neopeptides found in the tumor. These data support the hypothesis that the large proportion of mutant neoantigens in MMR-deficient cancers make them sensitive to immune checkpoint blockade, regardless of the cancers’ tissue of origin.