Profound neuroinflammation associated with long COVID

In a recent study posted to the medRxiv* pre-print server, researchers used a novel quantitative assessment, [¹⁸F]DPA-714 positron emission tomography (PET), to gather in vivo evidence of widespread neuroinflammation in two patients with post-acute sequelae of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection (PASC).

Study: Long COVID is associated with extensive in-vivo neuroinflammation on [18F]DPA-714 PET. Image Credit: DOERS / Shutterstock

Background

Around 36 to 53% of coronavirus disease 2019 (COVID-19) patients develop persistent chronic symptoms, often referred to as long COVID or PASC. These symptoms are primarily neurological and manifest as fatigue, dysgeusia, anosmia, and cognitive impairments.

Although long COVID has affected millions worldwide; however, relatively little is known about its pathophysiology. Studies have shown microglial activation and neuroinflammation in the brain of patients with long COVID. Therefore, more in vivo studies examining neuroinflammation in COVID-19 patients are needed. It would help elucidate the pathophysiological cascade underlying neurological symptoms of long COVID and help find potential remedial therapies.

About the study

In the present study, researchers enquired the two study patients to provide informed consent for thorough screening. Subsequently, they provided their medical records and participated in standardized neuropsychological assessment tests.

The first patient was a Dutch woman in her late fifties who worked full-time. She first contracted COVID-19 in December 2020, confirmed by a positive reverse transcription-polymerase chain reaction (RT-PCR). She did not need hospitalization or treatment during her acute phase of infection. However, she developed persistent fatigue, parosmia, anosmia, and some visual and concentration deficits later. Before the SARS-CoV-2 infection, she suffered from high cholesterol and fibromyalgia, but her condition was stable and did not interfere with her working. Following the infection, her fatigue worsened and persisted (along with other symptoms) for 15 months.

The second patient was a Dutch man in his mid-sixties who worked full-time. A positive RT-PCR indicated that he contracted SARS-CoV-2 infection in March 2020. During 15 days of hospitalization during the acute phase of SARS-CoV-2 infection, he stayed for one night in the intensive care unit (ICU) due to some respiratory problems. He faced difficulties in concentration and suffered from severe fatigue post-COVID-19. Since his symptoms persisted 24 months after infection, he was declared partially unfit for work. Both the patients were unvaccinated at the time of infection but received their vaccination after one year of SARS-CoV-2- infection. They received their respective primary vaccination series by one year and 21 months after infection.

The researchers determined the genotype of the RS6971 polymorphism in the translocator protein (TSPO) encoding gene. The three healthy control subjects of the current study matched with the study subjects in having a high affinity for TSPO. Notably, the first control subject was a female, while subjects two and three were males.

The researchers also included data from eight multiple sclerosis (MS) patients for (quantitative) comparison of the [ ¹⁸F]DPA-714 metabolism. They subjected all the study participants to magnetic resonance imaging (MRI), and also obtained their [¹⁸F]DPA-714  PET scans with arterial blood sampling.

Further, the researchers compared the [¹⁸F]DPA-714 metabolites in the blood of the two long COVID patients. Likewise, they compared brain gray matter BP_ND values between two long COVID patients and three matching healthy control subjects using the 2T4k_V_B  approach. Lastly, they generated volume-of-distribution (V_T) images using Logan plot analysis. Using t*=10 minutes, they divided those images by the whole brain grey matter k1/k2 ratio, following subtraction of one to correct the Logan V_T images for the non-displaceable distribution volume resulting in BP_ND (=k3/k4) images.

Neuroinflammation in two long COVID patients. To quantify [18F]DPA-714 binding in whole-brain gray matter (GM) we used a plasma input two tissue compartment model with blood volume parameter (2T4k_VB). All quantitative whole-brain grey matter binding potential (BPND (=k3/k4)) values reported are estimated using 2T4k_VB. For visualization purposes we generated volume-of-distribution (VT) images using Logan plot analysis (11), using t* = 10 min, and divided those images by the whole brain grey matter k1/k2 ratio obtained by the plasma input 2T4k_VB model, following subtraction of 1. By doing so, Logan VT images were corrected for the non-displaceable distribution volume resulting in BPND (=k3/k4) images for illustrative purposes. (A) T1-weighted MRI and parametric images of [18F]DPA-714 binding in the brain of two long COVID patients. Higher binding potential (BPND) values indicate more tracer binding and thus higher levels of neuroinflammation. Long COVID patient 1 showed severely elevated binding in all brain regions compared to the healthy control subjects. Long COVID patient 2 also showed elevated binding, with higher BPND values than the healthy control subjects. (B) T1-weighted MRI and parametric images of [18F]DPA-714 binding in the brain of three healthy control subjects.

Study findings

As per the neuropsychological test scores, both long COVID patients suffered fatigue, severe functional impairment, and concentration problems. The first patient had mildly impaired sustained attention and verbal memory deficits, whereas the second had fluctuating sustained attention and visuo-constructive deficits.

Compared to a [¹⁸F]DPA-714 cohort of healthy control subjects and MS patients, the tracer parent fraction and whole blood activity concentration corrected for both the study patients were within the range. Thus, the differences in tracer metabolism could not reasonably explain any differences in [¹⁸F]DPA-714 binding.

The MRI of the healthy control subjects and the first long COVID patient was consistent with age; however, the MRI of the second patient had mild atrophy in the parietal region. Furthermore, the first patient showed severely elevated [¹⁸F]DPA-714 binding in all brain regions. Compared to healthy controls, BP_ND (=k3/k4) values obtained from the 2T4k_V_B model in the first patient were increased by 121% on average, whereas the same values for the second patient increased on average by 79%.

Conclusions

The study data indicated widespread increases in [¹⁸F]DPA-714 binding throughout the brain in the two long COVID patients. The extent and magnitude of the study observations might not be definitive but were striking; thus, there is a dire need for further research to understand whether anti-inflammatory treatment could be beneficial for long COVID patients.

*Important notice

medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.