Michelle Broekhuizen and colleagues present a histopathological study comparing idiopathic and SARS-CoV-2 induced histiocytic intervillositis (HI) using multi-parametric immunohistochemical staining for spatial immune profiling. The authors clearly show how the intervillous spaces in SARS-CoV-2 antigen positive placentas gets infiltrated by myeloid cells, in particular pro-fibrotic monocytes, explaining the massive obstruction and fibrin deposition seen in SARS-CoV-2 placentitis. With this regards, SARS-CoV-2 placentitis is demonstrated to resemble chronic HI (CHI); with as only obvious difference relatively low expression of apoptosis markers.

The study is well designed and conducted. Apparently, such deep spatial proteomic analysis of this important pathology has not been reported before. Nevertheless, few minor questions remain (The manuscript lacks page and line numbering which makes a point-by-point critcism cumbersome). Presentation of data may benefit from a more detailed description of the actual methodology and approach for quantitative analysis.

Specific comments:
Page 3, §3: […] finding of SARS-CoV-2 placentitis, including chronic intervillositis and perivillous are also found in chronic histiocytic intervillositis (CHI).
P4 §3: The authors compare placentas of different gestational age (virus infected: 29w; controls 37w). This fairly large age difference may need to be discussed (whether or not this may skew the findings).
P4 §3: Figure 2E does not show a specimen from CHI, but during SARS-VoV-2 infection
P5 §1: The authors scored for M2 macrophages. Seen the virus-induced pathology, why was no specific staining for pro-inflammatory M1 macrophages (e.g. CD86) conducted. What could be expected? This may need some mention in results or discussion.
P5 §4: immune proteins
P6 discussion: The authors mention that VoC Delta caused frequently fetal damage. However, such a focus on later more aggressive vairiants is misleading as all specimen analyzed originate from 2020/21, thus prior to Delta.
P6 discussion: It is repeatedly mentioned that placentitis causes fetal damage due to impaired oxygen exchange, without providing experimental evidence. The study would benefit from a respective staining and quantitative analysis (e.g for HIF1a) to make the mechanistic link.
P7 §1: adaptation (instead of adaption)
P7 §2: A question remains, if placental insufficiency in SARS-CoV-2 infection requires direct virus-induced damage in the placenta. Technically speaking, is the IHC specimen shown 2E fully representative for all 9 placentas tested. How does the spatial immune profile look like of placentas that remain virus antigen negative while the mother suffers from lung infection? Likewise, what is the evidence that the here presented idiopathic CHI cases (that have also been sampled during the pandemic) are also not caused by inapparent/mild/passed maternal COVID? Thus “idiopathic” in lack of clinical or virological diagnosis?
P8 §2: The authors speculate that the pathogenesis of SARS-CoV-2 induced and idiopathic CIH may resemble that of haemophagic lymphohistiocytosis (HLH) regarding a local cytokine storm. The substantiate this hypothesis, the study may greatly benefit from a matching analysis. Detection of the “cytokine storm” should be feasible by spatial immune profiling.
P8 §4: See comment before; what happens in absence of placental infection. How does this compare to directly virus-induced immune pathogenesis?
P9 §1: haematopoietic
P9 §3: ROI, abbreviation not defined.
P9 §3: “additional mechanism”; here at least a mention of the lack of direct cytokine measurements (and M1 macrophages?) may deserve a mention.
P11 §1: Triplicates for each ROI were taken into account. How were these ROI selected? Which measures were taken to make sure that they are representative?
Same §: “The following protein panels were applied on the nCounter platform: […].” To the best of our knowledge nCounter is a digital transcriptomic platform and not an IHC technology. Please explain/correct what is meant here (Using the same gene ontologies as nCOunter uses?). Also, in Suppl Table 2 no specifics of the antibodies employed (e.g. clone, host, dilution, source + cat. no.) are mentioned. We think this is essential for reproducibility.
P11 §2: In the current description it is not clear how noise-to-background was managed. What were these “positive controls” and “negative controls”, what is Ms IgG2a? Also PanCK “was barely used” yet still shown in main Figure 1C. Considering that the authors conduct a quantitative analysis, their methodology for quantitative assessment should be less ambiguously described.
Figure legend 1C: Please specify in the legend what is represented by three markers PanCK, CD45 and CD68 used herein.
Figure D-F: The y-axes span 0-8 log10 while data span maximally 5 log10. I suggest to trim them for better visuability.
Figure 2: Scale bars missing.
Figure 3 and Suppl Table 1: The authors link MHC1 (represented by b2-microglobulin) expression functionally to antigen expression and tumor. We think both is not correct in the context of the placenta, nor in the context of infection. As MHC I are found on all nucleated body cells, what is the point of scoring for MHCI in the current study?
Figure 3: The study used n=9 specimen per group. However, only 5-7 data points are shown. This needs to be explained.
Figure 4: Item. Further, abbreviations LMM and BH not defined.

All my requests have been addressed to full satisfaction.
Many thanks to the authors for your thorough revision and thoughtful comments on our concerns. We hope that we could help improving your manuscript with our interventions.