Authors’ Response

We thank Drs. Glass, Omalu, and Hammer for their interest in our study. They raise several important questions regarding the neuropathological manifestations of chronic traumatic encephalopathy (CTE) and the importance of clinicopathological correlations. We are pleased to have the opportunity to discuss and clarify these points.

The Second NINDS Consensus Conference on CTE was done for three reasons: (i) to validate and refine the preliminary pathological criteria for CTE, provided by the First NINDS Consensus Conference [1], using a second blinded sample of cases of CTE representing all severities of disease, mild to severe; (ii) to develop the minimum threshold for CTE diagnosis; and (iii) to determine whether the McKee staging scheme for CTE [2,3] was reliable using a limited number of paraffin-embedded slides [4]. The Second NINDS Consensus Conference on CTE pathology was, by design, entirely histopathological. A separate series of NINDS-sponsored Consensus Conferences are currently underway to examine the provisional research diagnostic criteria for the clinical syndrome of CTE, known as Traumatic Encephalopathy Syndrome (TES), which were initially published in 2014. An assessment of the validity of the 2014 clinical research criteria for TES using CTE pathology as the gold standard was recently the focus of a large clinicopathological correlation study (n = 336) [5]. Cognitive symptoms, as opposed to mood, behavior, or motor symptoms, were associated with CTE pathology. This finding informed the revision of the TES research diagnostic criteria, published in 2021, which now require cognitive impairment to meet criteria for “possible” or “probable” TES [6]. An important clarification of the pathological criteria for CTE that was adopted by the second NINDS consensus panel is the necessity of neuronal tau in the CTE lesion [4].

Secondly, the panel pathologists were in large agreement about which cases represented CTE despite being blinded to the demographic, clinical and gross neuropathological features of the case. This speaks to the rigor of the study and precision of the proposed criteria. The fact that their agreement further improved after being provided demographic, clinical, and gross neuropathological information would be true for any neurodegenerative disease under evaluation, and it is particularly true when the trigger for the disease is overwhelmingly environmental, i.e. exposure to repetitive head impacts. In general, more information will almost certainly improve diagnostic accuracy regardless of the outcome (pathological, clinical, etc.).

We do not agree that “it is concerning that 2 cases were excluded because they ‘failed to include diagnostic p-tau pathology’, though CTE was diagnosed by the experts at BU.” The low stages of CTE are characterized by isolated small perivascular lesions in the cortex. All the slides provided for evaluation by the consensus panel were re-cut and re-stained for technical consistency. It is well recognized that small, focal lesions might not appear in full measure on recut slides and a brain with previously diagnosed low CTE might not show fully diagnostic perivascular lesions on re-cut slides through the same area.

The CTE cases were football players, although one was a boxer. Many CTE cases identified in the literature using the NINDS-NIBIB criteria have been amateurs and have played a variety of contact sports [7]. The intent of the study was not to identify differences in CTE pathology between players of different sports. In addition, no study to date has shown differences in CTE pathology based on primary exposure.

Drs. Omalu and Hammer raise several points regarding the pathological criteria for CTE. They question the validity of the NINDS-NIBIB criteria given that brainstem pathology is not considered essential to the diagnosis and they suggest that the pathological diagnosis of CTE, as defined by the NINDS-NIBIB criteria, does not consider the broader spectrum of acute and chronic pathologies that can occur after exposure to head trauma.

We are in full agreement with Drs. Omalu and Hammer that brainstem pathology is a feature of CTE, and the presence of p-tau related pathologies, including abnormal p-tau immunoreactive neuronal and astrocytic aggregates in the midbrain tegmentum and isodendritic core (nucleus basalis of Meynert, raphe nuclei, substantia nigra, and locus coeruleus) are considered supporting features of CTE by the First and Second Consensus panels [1,4]. The midbrain, pons, and medulla are also recommended regions to be sampled and evaluated [1,4] and the proposed diagnostic workflow includes evaluations of the brainstem in the diagnostic evaluation [4]. Brainstem p-tau pathology is a critical aspect of the McKee staging scheme [2] and a detailed examination of the regional distribution of p-tau in CTE including the brainstem, was recently quantitatively and semi-quantitatively assessed in 366 cases of CTE [3]. Furthermore, K-medoids cluster analysis of the semi-quantitative scales of p-tau across 14 key regions, including substantia nigra and locus cœruleus, identified 5 clusters of p-tau that conformed to increasing CTE stage, age at death, dementia, and years of American football play. There was a predilection for p-tau pathology in five regions: dorsolateral frontal cortex, superior temporal cortex, entorhinal cortex, amygdala, and locus coeruleus, with CTE in the youngest brain donors and lowest CTE stage restricted to dorsolateral frontal cortex and locus coeruleus.

The issue at hand in developing diagnostic criteria for CTE was determining what features were distinctive or unique to CTE that could be used to distinguish CTE from other tauopathies such as Alzheimer disease (AD) or progressive supranuclear palsy (PSP). Brainstem NFT and neurites, while common features of CTE, are not, in and of themselves, easily distinguishable from AD or PSP. All three disorders are characterized by NFTs and neurites in the substantia nigra, locus coeruleus and raphe nuclei, albeit there is greater regional involvement and neuronal loss in brainstem nuclei in PSP. In CTE, the pathology that distinguishes the disease from other tauopathies is the presence of a cortical pathognomonic lesion, not the presence or features of brainstem p-tau pathology.

Moreover, CTE is not a mutually exclusive disease, nor was it ever considered to be. The diagnostic entity CTE was never intended to represent the broad spectrum of all post-traumatic neurodegenerative pathologies. Other post-traumatic pathologies such as: beta amyloid deposition, alpha-synuclein immunopositive Lewy bodies, TDP 43 immunoreactive pathology, axonal loss, white matter degeneration, neuroinflammation, and microvascular alterations have been documented to occur after neurotrauma with and without co-morbid CTE. We support current NINDS efforts to further determine the broad spectrum of neurodegenerative pathologies that occur, acutely and chronically, after neurotrauma of all types and severities. CTE is one outcome; there are undoubtedly many others.