2024 Banff Meeting Report: Kidney

Thank you for the great effort of performing the post-meeting survey and the very helpful 2024 Report, which addresses a lot of recurrent questions that one is confronted with in daily routine.

There is some very small points I was wondering about:

-Figure 1

Diagnosis of chronic-active TCMR:

In the 2019 classification it was specifically stated that chronic-active tubulointerstital rejection is defined by i-IFTA2 or 3 AND ti2 or 3 with t2 OR tIFTA2 for Grade IA and with t3 OR t-IFTA3 for Grade IB. In the susequent publications, also the present one, it appears that t-IFTA2 or 3 without t2 or t3 would not be sufficient for the diagnosis. What was the reason to change this back? To me it appeared a sensible change of the definition.

List of concomitant causes of tubulointerstitial inflammation:

-        Neutrophilic tubulitis is also very helpful in the diagsnosis of ascending infection.

-        (IF/IHC/EM) instead of (IF/EM) might be considered

-Table S3

In the table of comments it says repeatedly that „Biopsy-based transcripts diagnostics could be considered if available“. In cases with findings below threshold for a rejection diagnosis this is surely very helpful. However, is there actually enough evidence to recommend this proceeding in cases, in which rejection cannot be differentiated from eg. TMA of other cause or GN. Can AMR or TCMR be differentiated from  TMA or GN using one of the currently used transcript tools? Sorry, if I am not up to date with the literature.

I understand this statement of table 2 should be rephraised to avoid "I would agree"

If clinical experience and research data support such major change, I would agree with eventually replacing ...

Table 4 includes 2 statements that are practically the same and may one of them be chosen

• Biopsy-based transcript diagnostics are included in the Banff classification of

  Category 2 AMR/MVI since more than 10 years. It is the task of the Banff

  classification to define/guide their context of use.

• Biopsy-based transcript diagnostics are included in the Banff classification of

  Category 2 AMR/MVI since more than 10 years, but their context of use remains

  relatively unclear in the Banff classification.

I suggest including this recent reference regarding advances in the study of the role of NK cells in AMR/MVI: Alloreactive adaptive natural killer cells in renal transplantation: Potential contribution to allograft microvascular inflammation. Alari-Pahissa, Elisenda et al.

American Journal of Transplantation, 2025, Volume 25, Issue 8, 1657 - 1669

REGARDING BERTRAND'S COMMENT; I am not sure how microarray based transcriptomics (= MMDx) can be recommended to comment on the presence of concurrent TCMR in a biopsy with an isolated v-lesion. By definition, these biopsies have virtually no interstitial inflammation. Since MMDX classifiers are driven primarily by intensity of inflammation, it is expected that the majority will get classified as not TCMR. The small proportion that do get called TCMR likely have inflammation only in the piece taken for molecular studies. It can be argued that this inflammation would have been generated a histologic TCMR call had the entire biopsy had been sent for routine histology. Histology is much more sensitive than MMDX for TCMR because of the focal nature of its lesions which may not be sampled by a 3mm piece taken for RNA isolation. Approximately, 50% of all histologic TCMRs that pathologists have been diagnosing and treating for the past 50 years do not get confirmed by MMDX

REGARDING MARION'S COMMENT:

BKV nephropathy, Pyelonephritis, drug induced interstitial nephrites, granulomatous disease, and glomerular disease with interstitial inflammation can all result in false positive signals that mimic the molecular TCMR signal

 Thank you very much for the excellent summary of the Banff 2024 meeting, the follow-up discussions, and the survey. Below are my comments:

• The new Figure 1 is very useful.

  Regarding the sentence “Evaluate TCMR threshold and/or biopsy-based transcripts for TCMR”: Have biopsy-based transcripts been studied separately in all differential diagnoses of TCMR? In other words, are we certain that the molecular signature of all the differential diagnoses of TCMR differs from the TCMR signature?

  Also, the threshold for Borderline could be added to align with the first box of TCMR.

  Abundant neutrophils in the infiltrate and medullary inflammation could also be suggestive of infectious disease.

• I have a concern regarding the “v” flowchart (Figure 2). It appears that when AMR criteria are met in the presence of a v lesion (first column), the diagnosis is always Mixed Rejection. Does this mean that AMR with a v lesion never exists on its own? Could we instead state that it is AMR with possible TCMR grade II or III if tubulointerstitial inflammation does not meet TCMR criteria, and probable Mixed Rejection if tubulointerstitial inflammation meets TCMR criteria?

  Moreover, in Table 1, it is stated that when two diagnoses are present together in the same biopsy, it is important to report each of them separately rather than using the term “Mixed Rejection.”

• Other remarks:

  • Abstract, line 10: Digital and molecular tools are discussed to help strengthen the Banff system for differential diagnostic reasoning, and to help transform the classification into a probabilistic tool reflective of the underlying immunological processes. I think non-invasive biomarker tools should also be mentioned in this context. The recent use of dd-cfDNA or urinary chemokines is also changing the reasoning in the interpretation of histological lesions.

  • Page 5, line 4: The Borderline category also excludes the acute TCMR category.

  • Page 17, line 13: Typo — should read “Supplementary Table S7” (and not “supplemantary”).

  • For activity and chronicity indexes: Could alternatives be proposed or discussed in Table 3?

    For example, the addition of the “ah lesion” in the chronicity index (very often seen and severe in late biopsies).

    Also, the addition of i-IFTA and/or ti in the activity scores should be considered.

    Different activity and chronicity indexes should be compared in future studies in order to find the more accurate.

Thank you for your work,

Sincerely,

Marion Rabant

Congratulations for this great work (Figures 1 and 2 are a great addition) and thank you for letting us the opportunity to comment it.

- Figure 1: 

=> Could we also say that inflammation within the medulla is more suggestive of infections or urological complications rather than rejection ? That edema is possibly more likely to be seen in rejection than PVN ?

=> some minor typos: you could standardize « etc » and « etc. » (in the urological line). The dot introducing the urological complications is not aligned nor of the same shape as the others. Probable missing space between « CRP testing » and « other ».  

- Figure 2:

=> MVI DSA- C4d- and v lesion is called mixed rejection. It seems to imply that MVI DSA- C4d- is always considered as rejection, and even AMR rejection, at least in the context of associated v. But the Banff 2022 report suggested other mechanisms. Quoting Table 3 of the 2022 report: « These cases may represent missed HLA-DSA, alloreactive T cell–mediated responses; autoreactive or alloreactive non-HLA antibodies; primary NK cell activation through missing self; viral infection; other mechanisms of innate immune activation; ischemia reperfusion injury, etc. ». Following Mr. Tsuji comment, it also implies that v is always TCMR-related, when it can also be part of the AMR diagnosis itself.

=> Is it possible to make a comment (or add a new part) about chronic allograft arteriopathy, where the reasoning shares some similarities with v+? It is a non rare finding.

=> missing middle arrow from the first to second block. Ideally, forms from the same line could be standardized in shape and aligned, which is not the case. In that way Figure 1 is a bit cleaner.

=> Possible double space between « No diagnosis of AMR/MVI » and « reached ».

- page 19: I understand that experts agreed on the potential usefulness of analyzing TCMR transcripts in context of « isolated v lesion ». Still, I wonder if the signal to noise ratio of a bulk approach is sufficient in this context of highly focal lesion, and I do not know if there are data supporting this. Depending on the experience of the community, a comment or note of caution about the interpretation could be added. Anyway, in the one or 2 cases that I have seen, it responded low to moderate probability of TCMR (so "molecularly insufficient" for a TCMR diagnosis).

- Table S3, second comment: « In sensitised patients early post-transplant, this indicates probable AMR. » => could we clarify if associated acute tubular injury is mandatory to retain probable AMR in this context? Even if early post-transplantation ATI is virtually always present, it is clearly stated in the reference guide but not here.

- Table S6/TCMR working group: points 1 and 5 are identical except the explanation of CCTT. Points 2 and 6 are identical. Points 3 and 8 are nearly identical except the last sentence of point 8 which is not part of this statement I believe.

- Sometimes ABMR is used instead of AMR (page 18, Table S6 and Table S13)

Again, congrats for this work,

Sincerely,

Bertrand Chauveau

Thank you for the invitation to speak at the Banff 2024 meeting and the opportunity to comment from a patient advocacy perspective.

1) While advancing knowledge and scientific understanding of AMR/MVI is promising, it is important to acknowledge the patient community's understanding of AMR/MVI. Based upon my advocacy work, I have learned that the majority of the patient community doesn't understand the concept of AMR. My recommendation is to conduct a survey of the global kidney transplant community on the subject of MVI/AMR. Included in that survey should be questions regarding post transplant monitoring regarding standard lab testing, DSA testing, cell free DNA, etc. It is my belief that a significant patient educational gap exists as well as standardization in post kidney transplant care.

2) I am reiterating my recommendation to form a Banff Foundation for Allograft Pathology global patient advisory board. Due to the composition of the foundation, a unique opportunity exists to strengthen the patient voice globally to advise the foundation, advocate for additional research and pharma investment, and improve implementation through their patient organizations.

Sincerely,

Kevin J. Fowler

Principal, The Voice of the Patient

kevinjohnfowler@gmail.com

Thank you for publishing the Banff 2024 Meeting Report—I had been looking forward to its release. The flowcharts in Figures 1 and 2 are extremely helpful for understanding the complexities of rejection diagnostics.

Two minor suggestions for Figure 2

All three branches that follow “v lesions present” run in parallel. To make that parallel structure more intuitive, it may help to add a direct arrow from the parent node to the central box (“Tubulo-interstitial inflammation at or above t1/i1 …”) just as you have arrows to the left- and right-hand boxes.

The lower-left box is labeled “Mixed rejection (AMR/MVI + TCMR grade II)”. Because biopsies with a v3 score also follow this branch, “TCMR grade II or III” would be more accurate.

Question on interpretation

The same lower-left box classifies AMR/MVI cases with a v lesion as “Mixed rejection (AMR + TCMR)”. Yet intimal arteritis (v lesions) can arise within the pathophysiology of AMR itself. Might that box therefore encompass both pure AMR and Mixed rejection, depending on the broader context? I would appreciate the Working Group’s perspective on how the v lesion is weighed when deciding whether TCMR is truly concomitant.

Thank you again for making this valuable report available. I look forward to further discussion and future developments.

Sincerely,

Takahiro Tsuji