It depends entirely on what you mean by peer review. Using Australia's Higher Education Research Data Collection (HERDC) criteria for the collection of publication information, peer review amounts to substantial review by appropriate experts. Books are either reviewed by proxy by commercial publication in this research data collection scheme, reliant on the habit of scholarly monographs being sent out for readers and scholarly collections being edited by the book's editors, or they have to be explicitly peer reviewed if published in an electronic only format. The quality of review of monographs varies considerably from field to field and from publishing house to publishing house. It matters quite a great deal for my discipline as my discipline is monograph driven.
Book reviews are sometimes peer reviewed. Book reviews of more than one work which advance original scholarly arguments are regularly peer reviewed in my domain (HASS: Humanities, Arts, Social Sciences; the converse of STEM). These reviews contribute either a critical reflection on current practice, or they contribute a substantive account of the field (literature review / review article / field review). HERDC recognises this to the extent that reviews of more than one work which otherwise meet the criteria of a journal article (contribution to scholarly knowledge, peer review) are accepted as full journal articles.
Some news articles in journals formally meet the standard for peer review of HERDC, but fail other criteria, such as an original contribution to scholarly knowledge.