Abstract
Background
Neurofilament light (NFL) chain concentrations, reflecting axonal damage, are seen in several polyneuropathies but have not been studied in human immunodeficiency virus (HIV) distal sensory polyneuropathy (DSP). We evaluated NFL in cerebrospinal fluid (CSF) and plasma in relation to DSP in people with HIV (PWH) from 2 independent cohorts and in people without HIV (PWoH).
Methods
Cohort 1 consisted of PWH from the CHARTER Study. Cohort 2 consisted of PWH and PWoH from the HIV Neurobehavioral Research Center (HNRC). We evaluated DSP signs and symptoms in both cohorts. Immunoassays measured NFL in CSF for all and for plasma as well in Cohort 2.
Results
Cohort 1 consisted of 111 PWH, mean ± SD age 56.8 ± 8.32 years, 15.3% female, 38.7% Black, 49.6% White, current CD4+ T-cells (median, interquartile range [IQR]) 532/µL (295, 785), 83.5% with plasma HIV RNA ≤50 copies/mL. Cohort 2 consisted of 233 PWH of similar demographics to PWH in Cohort 1 but also 51 PWoH, together age 58.4 ± 6.68 years, 41.2% female, 18.0% Black, Hispanic, non-Hispanic White 52.0%, 6.00% White. In both cohorts of PWH, CSF and plasma NFL were significantly higher in both PWH with DSP signs. Findings were similar, albeit not significant, for PWoH. The observed relationships were not explained by confounds.
Conclusions
Both plasma and CSF NFL were elevated in PWH and PWoH with DSP. The convergence of our findings with others demonstrates that NFL is a reliable biomarker reflecting peripheral nerve injury. Biomarkers such as NFL might provide, validate, and optimize clinical trials of neuroregenerative strategies in HIV DSP.
Keywords: HIV, polyneuropathy, biomarker, cerebrospinal fluid, neurofilament light
In 2 independent cohorts, levels of neurofilament light, a marker of axonal degeneration, in blood plasma and cerebrospinal fluid were elevated in people with human immunodeficiency virus (HIV) who demonstrated exam findings of distal sensory polyneuropathy.
We and others have demonstrated that distal sensory polyneuropathy (DSP) is a disabling chronic condition in people with human immunodeficiency virus (HIV, PWH),—even with viral suppression on antiretroviral therapy (ART) [1]. Consequences of DSP include disability [2, 3], neuropathic pain [1], and poor balance [4, 5]. Neuroregenerative strategies are becoming available [6]. A reliable surrogate marker for peripheral nerve regeneration might make clinical trials of neuroregenerative strategies more efficient and provide biological validation [7, 8]. Cerebrospinal fluid (CSF) neurofilament light (NFL) might be a useful marker [9–13], as CSF ensheaths the proximal nerve root sleeve [14], and thus its molecular constituents, including biomarkers of axonal injury, likely reflect dorsal root ganglia (DRG) health [15]. HIV-associated DSP is recognized as a DRG sensory neuronopathy, likely caused by the infiltration of activated macrophages into DRG [16–18] [19]. Cerebrospinal fluid (CSF) and serum NFL levels also are increased in a variety of peripheral neuropathies [11–13, 15, 20–22], but this has not been systemically evaluated in HIV DSP. We tested the hypothesis that participants with DSP would have higher CSF and plasma NFL levels than those without DSP.
METHODS
Study Participants
We studied 2 independent, prospective, community-based cohorts of PWH. Cohort 1 was a multicenter (CNS HIV AntiRetroviral Effects Research [CHARTER] [1], March 2016 to January 2020) and Cohort 2 was a single-center (San Diego HIV Neurobehavioral Research Center [HNRC] [23], October 1999 to March 2019). All participants in Cohort 1 were HIV seropositive; Cohort 2 included both HIV seropositive and HIV seronegative individuals. Inclusion criteria were HIV seropositive or HIV seronegative and having had a systematic evaluation for DSP signs and symptoms. Exclusions were individuals with a history of diabetes mellitus preceding HIV, neurological illnesses unrelated to HIV that might confound the study assessments, and inability to comply with the study evaluations. Both studies were approved by local Institutional Review Boards; the experiments were undertaken with the understanding and written consent of each subject; the study conformed with the World Medical Association Declaration of Helsinki.
DSP Evaluations
Participants in both cohorts were evaluated by trained clinicians using standardized clinical examinations for findings of DSP and interviews for symptoms of DSP. DSP signs were bilateral, distal vibration, sharp, and ankle reflex loss. Participants were also categorized as having 0, 1, or ≥2 DSP signs. Symptoms were neuropathic pain, paresthesias, and loss of sensation. To avoid the inclusion of mononeuropathies and radiculopathies, in the case of both signs and symptoms, we required symmetrical, bilateral, distal abnormalities.
Other Clinical Evaluations
A trained clinical examiner interviewed and examined participants to collect information such as ART regimens, nadir CD4+ T-cell counts, and history of diabetes mellitus. Although this study was not focused on the relationship of NFL to neurocognitive impairment, we nevertheless evaluated the latter as a potential confound. Neurocognition was measured using a comprehensive neuropsychological battery covering seven domains as specified by the Frascati criteria [24]. The battery is described in detail in a previous publication and included tests of executive function, learning, memory, attention, working memory, psychomotor speed, and speed of information processing [25]. Raw test scores were converted to standardized T-scores (mean 50, standard deviation 10) corrected for age, education, sex, and race/ethnicity, and overall performance was indexed by the global deficit score (GDS) method [26].
Clinical Laboratory Evaluations
All participants provided blood, and in Cohort 2, we collect CSF via lumbar puncture. HIV infection was diagnosed using an enzyme-linked immunosorbent assay (ELISA) with Western blot confirmation. HIV RNA was measured using commercial assays and considered undetectable at a lower limit of quantification (LLQ) of 50 copies/mL. CD4+ T cells were measured by flow cytometry and nadir CD4+ T-cell count by self-report.
NFL Assays
In Cohort 1, NFL was measured in duplicate in CSF only using a commercially available ELISA (TECAN Life Sciences; lifesciences.tecan.com). In Cohort 2, NFL was quantified in plasma and CSF using the ultrasensitive single molecule array (Simoa) platform using the Quanterix NF-light Advantage kit (no. 103186) at 1:4 dilution for plasma and 1:100 dilution for CSF, as recommended by the manufacturer. The intra-assay variability of the duplicate measurements was <5%.
Statistical Analyses
Demographic and clinical characteristics were summarized using means and standard deviations (SD), medians, and interquartile ranges (IQR) or percentages, as appropriate. NFL concentrations were log10-transformed to improve their skewed distribution. NFL levels in CSF and plasma were compared in participants with and without DSP signs and symptoms using analysis of variance (ANOVA) when the distribution of the outcome variable was not significantly different from normal. Non-parametric analysis was applied when variable distributions significantly deviated from normal. Pairwise comparisons were corrected using Tukey HSD test. To limit potential for Type 1 error, primary analyses were done for the relationship between number of DSP signs and NFL levels; all other analyses were secondary. Multivariable regression assessed the associations between NFL levels in CSF and plasma and DSP signs and symptoms after adjusting for relevant covariates and to test interaction effects between NFL and significant covariates. Because previous studies have shown that older nucleoside reverse transcriptase inhibitor antiretrovirals (so-called “d-drugs”) have substantial peripheral neurotoxicity, we evaluated this as another potential confounding condition. Analyses were conducted using JMP Pro version 15.0.0 (SAS Institute, Cary, North Carolina, USA, 2018).
RESULTS
Cohort 1 (PWH Only)
Demographics and HIV Disease Characteristics
Participants in Cohort 1 were 111 PWH, age mean ± SD 56.8 ± 8.32 years, 15.3% female, 38.7% Black, 10.8% Hispanic, 49.6% non-Hispanic White, estimated duration of HIV (median, IQR) 22.4 (16.9, 27.7) years, current CD4+ T-cells (median, IQR) 532/µL (295, 785), nadir CD4+ T-cells 95 (15, 199), 98.2% on ART, 83.5% with plasma HIV RNA ≤50 copies/mL, 90.9% with CSF HIV RNA ≤50 copies/mL and d-drug exposure median (IQR) 18.3 (0, 67.0) months. Participants took 51 different ART regimens, with the most common being dolutegravir/abacavir/lamivudine (12.8%), EFV/FTC/TFV (11.0%), COBI/EVG/FTC/TAF (9.17%), and DTG/FTC/TAF (6.42%). The most common regimen types were: integrase inhibitor + 2 nucleoside reverse transcriptase inhibitors (NRTIs) (38.5%), protease inhibitor + non-nucleoside RTI + 2 NNRTIs (20.2%), and 2 NRTI + NNRTI (19.3%).
NFL Levels According to DSP Signs, Cohort 1
Figure 1 shows that higher log10 CSF NFL was related to increasing numbers of DSP signs (ANOVA P = .0108); ≥2 DSP signs (N = 53, 3.15 ± 0.174), intermediate with 1 sign (N = 29, 3.13 ± 0.249), and lowest with 0 signs (N = 29, 3.01 ± 0.190). CSF NFL was significantly related to reduced vibratory sensation (3.14 ± 0.168 vs 3.054 ± 0.221, P = .016) and reflexes (3.17 ± 0.203 vs 3.02 ± 0.183, P = .0001), but not to reduced sharp sensation (3.11 ± 0.190 vs 3.09 ± 0.205, P = .619). CSF NFL was not related to DSP symptoms (neuropathic pain, paresthesias, loss of sensation; ps >0.50).
Figure 1.
Box plots showing CSF NFL levels in Cohort 1 according to the number of DSP signs. Boxes represent interquartile ranges, whiskers represent 5th (lower) and 95th (upper) percentiles. Cohen's d, effect size. Abbreviations: CSF, cerebrospinal fluid; DSP, distal sensory polyneuropathy; NFL, neurofilament light.
Potential Confounding Conditions, Cohort 1
Older age correlated with higher CSF NFL (r = 0.423, P = 3.64 x 10−6). Additionally, there was a stepwise increase in age according to the number of DSP signs: for ≥2 signs versus 1 sign versus 0 signs, age in years was 60.3 ± 7.44, 54.7 ± 6.97, and 52.4 ± 8.52, respectively (P = 2.68e−6). We therefore assessed a multivariable regression model that included age, DSP and their interaction as predictors of CSF NFL. Since the interaction was not significant (P = .975), it was removed from the model. We found the significant association between DSP signs and CSF NFL became non-significant (P = .165) after controlling for age (P = .00007). ART status (on vs off) was not related to log CSF NFL levels (P = .731) or to DSP signs (P = .495). D-drug exposure was not related to CSF NFL levels (P = .288) or to DSP signs (P = .678).
CSF NFL was highest in PWH of “Other” ethnicities (N = 1, 3.70) compared to non-Hispanic White, (3.09 ± 0.184) and Hispanic (3.06 ± 0.147) (P = .0187). In a multivariable regression, both the number of DSP signs and race/ethnicity, but not their interaction, were significantly associated with higher CSF NFL (ps = 0.00779 and .0134, respectively). Neurocognitive performance (GDS) was not significantly related to CSF NFL levels (r = −0.153, P = .108). CSF NFL levels were similar in participants with detectable versus undetectable plasma HIV RNA (3.12 ± 0.137 vs 3.11 ± 0.218, P = .758). CSF NFL was not related to CSF HIV RNA (r = 0.0305, P = .401) or to nadir (r = −0.124, P = .193) or current CD4+ T-cell count (r = −0.0731, P = .448), CSF NFL was not associated with diabetes mellitus (22.1% of participants), hepatitis C virus positive serology (42.3%), lifetime alcohol use disorder (55.0%), body mass index or serum creatinine (ps > 0.15). Total duration of past exposure to d-drugs (median 2.0, IQR 0, 36.9 months) was not related to CSF NFL (r = 0.092, P = .335) or to DSP signs (ANOVA P = .617). Currently taking ART also was not related to CSF NFL (P = .731) or to DSP signs (P = .495). ART regimen and regimen type were not related to CSF NFL or to DSP signs (ps > 0.15).
Cohort 2 (PWH and PWoH)
Demographics and disease characteristics for Cohort 2 are listed in Table 1. All PWH had undetectable plasma HIV RNA. PWH were much more likely to have ≥2 signs of DSP than PWoH (53.0% vs 6.0%, P = 1.01e−10). Among PWH, 33% had been exposed to d-drugs in the past.
Table 1.
Comparison of Demographics and HIV Disease Characteristics in Cohort 2
| All | PWH | PWoH | P | |
|---|---|---|---|---|
| N | 249 | 199 | 50 | … |
| Age in y—mean (SD) | 57.5 6.70 | 57.3 6.68 | 58.3 6.74 | .732 |
| Sex female—N (%) | 55 (22.1%) | 34 (17.1%) | 21 (42.0%) | .0015 |
| Race/ethnicity Black—N (%) | 44 (17.8%) | 35 (17.8%) | 9 (18.4%) | .629* |
| Race/ethnicity Hispanic—N (%) | 17 (6.91%) | 14 (17.1%) | 3 (6.1%) | … |
| Race/ethnicity non-Hispanic white—N (%) | 137 (55.7%) | 111 (56.3%) | 26 (53.1%) | … |
| Race/ethnicity other—N (%) | 48 (19.5%) | 37 (18.7%) | 11 (22.5%) | … |
| Nadir CD4—median (IQR) | … | 54 (13, 200) | … | … |
| Current CD4—median (IQR) | … | 453 (286, 668) | … | … |
| Undetectable plasma HIV RNA—N (%) | … | 199 (100%) | … | … |
| Est. duration of HIV in y—median (IQR) | … | 17.1 (11.7, 23.5) | … | … |
| On ART—N (%) | … | 199 (100%) | … | … |
| Duration of d-drug use—median (IQR) | … | 0 (0, 31.1) | … | … |
Bold values are statistically significant.
Abbreviations: ART, antiretroviral therapy; HIV, human immunodeficiency virus; IQR, interquartile range; PWH, people with HIV; PWoH, people without HIV; SD, standard deviation.
*P value for difference between PWH and PWoH across all ethnicities.
NFL Levels According to HIV Serostatus and DSP Signs, Cohort 2
Plasma NFL was significantly higher in PWH than PWoH (1.12 ± 0.275 vs 1.03 ± 0.208, P = .0321), and CSF NFL was numerically higher in PWH than PWoH (2.89 ± 0.297 vs 2.82 ± 0.244, P = .118). NFL was on average higher in CSF than plasma in both PWH and PWoH (for PWH, 2.7-fold higher, 2.89 ± 0.297 vs 1.12 ± 0.275; for PWoH, 2.6-fold 2.82 ± 0.244 vs 1.03 ± 0.208). CSF and plasma NFL levels were highly correlated in both PWH (r = 0.469, P = 2.77e−12) and PWoH (r = 0.527, P = 8.40e−5). Figure 2 shows that both CSF and plasma NFL levels were higher in PWH with abnormal DSP signs (for CSF, ≥2 signs 2.94 ± 0.302, 1 sign 2.82 ± 0.291, 0 signs 2.94 ± 0.302, ANOVA P = .0303; for plasma, >2 signs 1.17 ± 0.296, 1 sign 1.09 ± 0.229, 0 signs 1.05 ± 0.258, ANOVA P = .0309). Table 2 shows that among the three individual DSP exam findings, vibratory sensation was related to higher CSF NFL, and both vibratory sensation and reflexes were related to higher plasma NFL levels in PWH. Among PWoH, more DSP signs were associated with higher plasma NFL (≥2 signs 1.28 ± 0.0682 vs 1 sign 1.11 ± 0.177 vs 0 signs 0.970 ± 0.202, P = .0061), but not CSF NFL (≥2 signs 3.00 ± 0.204 vs 1 sign 2.82 ± 0.300 vs 0 signs 2.81 ± 0.216, P = .415).
Figure 2.
CSF and plasma NFL levels according to the number of DSP signs for PWH in Cohort 2. Boxes represent interquartile ranges, whiskers represent 5th (lower) and 95th (upper) percentiles. Cohen's d, effect size. Abbreviations: CSF, cerebrospinal fluid; DSP, distal sensory polyneuropathy; NFL, neurofilament light; PWH, people with human immunodeficiency virus.
Table 2.
CSF and Plama NFL Levels According to Each Sign of DSP for PWH (Cohort 2)
| CSF NFL | Plasma NFL | |||||
|---|---|---|---|---|---|---|
| Symptom/Finding | Normal | Abnormal | P | Normal | Abnormal | P |
| Vibratory sensation | 2.83 ± 0.0316 | 2.94 ± 0.0273 | .0057 | 1.05 ± 0.254 | 1.18 ± 0.280 | .0018 |
| Reflexes | 2.87 ± 0.264 | 2.91 ± 0.317 | .448 | 1.07 ± 0.248 | 1.16 ± 0.288 | .0360 |
| Sharp sensation | 2.88 ± 0.303 | 2.91 ± 0.284 | .574 | 1.12 ± 0.263 | 1.12 ± 0.301 | .988 |
Bold values are statistically significant.
Abbreviations: CSF, cerebrospinal fluid; DSP, distal sensory polyneuropathy; NFL, neurofilament light; PWH, people with human immunodeficiency virus,
Potential Confounds, Cohort 2
Among PWH, the number of abnormal DSP signs increased with increasing age (ANOVA P < .0013). CSF and plasma NFL levels also were higher in older participants (r = 0.300, P = 1.85e−5 and r = 0.291, P = 3.40e−5, respectively). In a multivariable model including the number of DSP signs, age, and their interaction as predictors of CSF NFL, age was significant (P = .00058), but DSP signs (P = .137) and the interaction of age with DSP signs (P = .930) were not significant. Similarly, in a multivariable model including the number of DSP signs, age, and their interaction as predictors of plasma NFL, age was significant (P = .0109); DSP signs (P = .174) and the interaction were not (P = .425). To further characterize these interrelationships, we evaluated collinearity. The variance inflation factor (VIF) for the collinearity of age and DSP signs was 1.07, indicating no collinearity.
Males and females did not differ with respect to CSF NFL (2.90 ± 0.291 vs 2.82 ± 0.326, P = .154) or plasma NFL (1.13 ± 0.276 vs 1.08 ± 0.279, P = .405). Among PWH, ethnicity was not significantly related to NFL in CSF or plasma (P-values 0.337 and 0.314, respectively). CSF NFL was not related to nadir (r = 0.0670, P = .350) or current (r = 0.00388, P = .958) CD4+ T lymphocytes. Similarly, plasma NFL was not related to nadir (r = −0.0551, P = .442) or current (r = −0.128, P = .0790) CD4 count. Duration of d-drug exposure was not related to DSP signs or CSF or plasma NFL levels (ps > 0.20). PWH with diabetes had more signs of DSP (no signs 14.6%, 1 sign 17.1%, ≥2 signs 68.3%) compared to those without diabetes (no signs 25.3%, 1 sign 28.6%, ≥2 signs 46.1%; overall P = .0386). Those with diabetes had similar levels of CSF NFL and plasma NFL (ps > 0.15) to those without. In PWH, neurocognitive performance (GDS) was not related to CSF or plasma NFL (ps > 0.10). There were no interactions between neurocognitive performance and HIV status for either CSF or plasma NFL (ps > 0.20).
Rates of the following comorbid conditions did not differ between PWH and PWoH: diabetes (21.0% vs 18.0%, P = .632); hepatitis C virus positive serology (28.2% vs 28.0%, P = .979). Higher log10 serum creatinine levels were associated with higher CSF NFL (r = 0.411, P = 2.13e−11) and plasma NFL (r = 0.528, P = 5.73e−19). In a multivariable model for CSF NFL, DSP signs remained significant (P = .0142), while creatinine and the interaction were not. In a multivariable model for plasma NFL, the interaction and the main effect of DSP signs were not significant, while creatinine was (P = 1.67e−16). Hepatitis C virus serostatus was not significantly associated with DSP (P = .879).
DISCUSSION
In 2 independent cohorts, PWH with abnormal signs of DSP had higher NFL levels in CSF than those without DSP, consistent with axonal injury in DSP. In Cohort 2, plasma NFL also was measured and showed higher levels with DSP than without. The observed relationships between DSP and NFL were robust to consideration of potential confounds including viral suppression, nadir and current CD4, diabetes mellitus, and use of neurotoxic d-drugs and ART. Although the relationship of NFL to CNS disease (cognitive impairment) was not the focus of these analyses, we found that cognitive impairment did not significantly confound the relationship between NFL and DSP. In both cohorts, the association of DSP with CSF NFL was no longer significant after adjustment for age. This finding suggests that aging may drive both neuropathy and elevated NFL. Elevated plasma NFL in those with DSP was not specific to PWH, but present also in PWoH, suggesting a more general mechanism of neuropathogenesis, rather than one particular to HIV DSP. NFL levels were not associated with DSP symptoms, suggesting the symptoms depend on factors other than neurodegeneration.
The frequency of DSP, defined as at least two neuropathy signs, was very high despite viral suppression. This, along with prior evidence that DSP continues to develop de novo and worsen in the modern era, particularly in older PWH [1], is consistent with an active peripheral neurodegenerative process. Among the potential sources of ongoing peripheral neurodegeneration are persistent neuroinflammation [27, 28], neurotoxicity of ART [29] and comorbidities [30], though the comorbidities evaluated here were not associated with DSP.
In both cohorts, higher NFL levels were seen with exam findings of larger fiber (vibration, reflexes), rather than small (sharp sensation) DSP. HIV DSP is a mixed large and small fiber axonal neuropathy [31]; our findings suggest that large fibers contribute to axonal injury to a greater extent than small fibers. This is plausible since the larger the caliber of the axons, the more microtubules—which are stabilized by neurofilament proteins—are available to degenerate, releasing NFL into the extracellular space, where it can be measured.
These findings, in conjunction with elevated CSF NFL in other peripheral neuropathies [20–22], raise the question of why peripheral nerve injury leads to elevated NFL in CSF as well as plasma. Levels in CSF were approximately 1000-fold higher than plasma, making it implausible that NFL passively diffuses from blood into CSF. We suspect that elevated CSF NFL in the context of peripheral neuropathies is related to the presence of CSF in nerve root sleeves surrounding dorsal root ganglia (DRG) [14], with elevated CSF NFL levels reflecting injury to DRG sensory neurons. Indeed, HIV DSP is recognized as a DRG sensory neuronopathy, likely caused by the infiltration of activated macrophages into DRG [16–18]. Consistent with this, in vitro, DRG neurons exposed to supernatants from HIV-infected macrophages showed axonal retraction without neuronal cell death [19].
Although HIV is known to cause DSP independent of other factors, HIV DSP is typically multifactorial; d-drug exposure and HIV-associated diabetes are particularly common in HIV and are believed to be important contributors. We did not exclude participants in whom these contributors existed. In fact, the inclusion of such cases enhances the generalizability of our findings to other cohorts. The neurotoxicity of older nucleoside reverse transcriptase inhibitor antiretrovirals (so-called d-drugs) is well-described, and indeed, although none of our participants took d-drugs at the time of assessment, many had been exposed to d-drugs in the past. However, neither diabetes nor d-drugs were related to NFL levels or DSP.
Strengths of this study include the replication across two independent cohorts that differed substantially in demographics, emphasizing the robustness and generalizability of our findings. Additionally, although we used different assay methods for NFL in the two cohorts, findings were similar for both assays.
Limitations of this study include its cross-sectional nature, precluding causal inference. Unobserved variables may have confounded our results. Reverse causation—elevated NFL causing peripheral neuropathy—is implausible. Neuropathy diagnoses were based on clinical exam findings; we did not perform electromyography and nerve conduction studies to further evaluate this. However, we have previously published on the predictive value of the clinical diagnosis using the approach described in the current manuscript compared to diagnoses using QST and NCV measures. In 1 study [32], the specificity of the clinical diagnosis of DSP was high (89.5%), and the positive predictive value was 84.6%. In a second study using similar methods [33], the clinical DSP correct classification rate relative to an electrophysiological approach was 78%, with a specificity of 88%. Thus, we are confident that our clinical diagnoses represent a high probability of DSP. Some neuropathy etiologies were not systematically evaluated here and could have been excluded with blood tests, strengthening the study. It was not practical to perform an exhaustive search for the myriad alternative causes, and this would have been cost-prohibitive. The number of female participants was relatively small, although no sex interactions were observed for the relationship between NFL and DSP. NFL levels in those with and without DSP substantially overlapped, limiting the usefulness of NFL for diagnostic purposes. Because NFL levels overlapped between the DSP and non-DSP groups, and because this study was not longitudinal, we cannot claim NFL as a diagnostic or prognostic tool, although the latter might be evaluated in future longitudinal studies.
The convergence of our findings in PWH with those in other polyneuropathies provides convincing evidence that NFL is a reliable correlate of peripheral nerve injury [11–13, 15, 20]. Notably, new neuroregenerative interventions have become available to treat DSP [6]. This is important because biomarkers such as NFL might be used as a surrogate marker of treatment response with neuroregenerative treatments, potentially increasing the efficiency of clinical trials [7, 8]. An additional future direction is considering whether elevated NFL levels might presage new onset DSP, offering possibilities for prevention if effective neuroprotective strategies become available.
Notes
Acknowledgments. The CNS HIV Anti-Retroviral Therapy Effects Research was supported by awards N01 MH22005, HHSN271201000036C, HHSN27120100003°C, and R01 MH107345 from the National Institutes of Health.
The CNS HIV Anti-Retroviral Therapy Effects Research (CHARTER) group is affiliated with Johns Hopkins University; the Icahn School of Medicine at Mount Sinai; University of California, San Diego; University of Texas, Galveston; University of Washington, Seattle; Washington University, St. Louis; and is headquartered at the University of California, San Diego and includes: Directors: Robert K. Heaton, PhD, Scott L. Letendre, MD; Center Manager: Donald Franklin Jr.; Coordinating Center: Brookie Best, PharmD, Debra Cookson, MPH, Clint Cushman, Matthew Dawson, Ronald J. Ellis, MD, PhD, Christine Fennema Notestine, PhD, Sara Gianella Weibel, MD, Igor Grant, MD, Thomas D. Marcotte, PhD, Jennifer Marquie-Beck, MPH, Florin Vaida, PhD; Johns Hopkins University Site: Ned Sacktor, MD (PI), Vincent Rogalski; Icahn School of Medicine at Mount Sinai Site: Susan Morgello, MD (PI), Letty Mintz, NP; University of California, San Diego Site: J. Allen McCutchan, MD (PI); University of Washington, Seattle Site: Ann Collier, MD (Co-PI), and Christina Marra, MD (Co-PI), Sher Storey, PA-C; University of Texas, Galveston Site: Benjamin Gelman, MD, PhD (PI), Eleanor Head, RN, BSN; and Washington University, St. Louis Site: David B Clifford, MD (PI), Mengesha Teshome, MD.
The views expressed in this article are those of the authors and do not reflect the official policy or position of the United States Government.
The HIV Neurobehavioral Research Center (HNRC) is supported by Center award P30MH062512 from NIMH. The San Diego HIV Neurobehavioral Research Center [HNRC] group is affiliated with the University of California, San Diego, the Naval Hospital, San Diego, and the Veterans Affairs San Diego Healthcare System, and includes: Director: Robert K. Heaton, PhD, Co-Director: Igor Grant, MD; Associate Directors: J. Hampton Atkinson, MD, Ronald J. Ellis, MD, PhD, and Scott Letendre, MD; Center Manager: Jennifer Iudicello, PhD; Donald Franklin Jr.; Melanie Sherman; NeuroAssessment Core: Ronald J. Ellis, MD, PhD (PI), Scott Letendre, MD, Thomas D. Marcotte, PhD, Christine Fennema-Notestine, PhD, Debra Rosario, MPH, Matthew Dawson; NeuroBiology Core: Cristian Achim, MD, PhD (PI), Ana Sanchez, PhD, Adam Fields, PhD; NeuroGerm Core: Sara Gianella Weibel, MD (PI), David M. Smith, MD, Rob Knight, PhD, Scott Peterson, PhD; Developmental Core: Scott Letendre, MD (PI), J. Allen McCutchan; Participant Accrual and Retention Unit: J. Hampton Atkinson, MD (PI) Susan Little, MD, Jennifer Marquie-Beck, MPH; Data Management and Information Systems Unit: Lucila Ohno-Machado, PhD (PI), Clint Cushman; Statistics Unit: Ian Abramson, PhD (PI), Florin Vaida, PhD (Co-PI), Anya Umlauf, MS, Bin Tang, MS.
Author Contributions. R. J. E.: Conceptualization, Funding acquisition, Formal Analysis, Writing—original draft
A. C., Y. L., D. C., J. W., B. T., C. M. M., L. H. R., D. B. C., J. A. M., B. B. G., J, R. P., C. J. P.: Investigation, Writing—review and editing
S. L, L.: Funding acquisition, Investigation, Writing—review and editing
Data sharing. Data will be made available on request to the corresponding author.
Financial support. This work was supported by the National Institutes of Health (grant numbers R01MH107345 to R. K. H. and P30MH62512 to R.K. H.). B, B. G. reports support from NIH (grant number N01 MH22005). L. H. R. reports support from NIMH. J. W., Y. L., D. C., and C. J. P. reports support from Monogram Biosciences, a Labcorp specialty testing group (as employee).
Contributor Information
Ronald J Ellis, Department of Neurosciences, University of California, San Diego, San Diego, California, USA.
Ahmed Chenna, Monogram Biosciences, South San Francisco, California, USA.
Yolanda Lie, Monogram Biosciences, South San Francisco, California, USA.
Dusica Curanovic, Monogram Biosciences, South San Francisco, California, USA.
John Winslow, Monogram Biosciences, South San Francisco, California, USA.
Bin Tang, Department of Psychiatry, University of California, San Diego, San Diego, California, USA.
Christina M Marra, Deparment of Neurology, University of Washington, Seattle, Washington, USA.
Leah H Rubin, Department of Neurology, Johns Hopkins University, Baltimore, Maryland, USA.
David B Clifford, Department of Neurology, Washington University at St. Louis, St. Louis, Missouri, USA.
J Allen McCutchan, Department of Medicine, University of California San Diego, San Diego, California, USA.
Benjamin B Gelman, Department of Neuroscience and Cell Biology, UTMB, Galveston, Texas, USA.
Jessica Robinson-Papp, Department of Neurology, Icahn School of Medicine at Mt. Sinai, New York, New York, USA.
Christos J Petropoulos, Monogram Biosciences, South San Francisco, California, USA.
Scott L Letendre, Departments of Medicine and Psychiatry, University of California, San Diego, San Diego, California, USA.
References
- 1. Ellis RJ, Diaz M, Sacktor N, et al. Predictors of worsening neuropathy and neuropathic pain after 12 years in people with HIV. Ann Clin Transl Neurol 2020; 7:1166–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Marcus KS, Kerns RD, Rosenfeld B, Breitbart W. HIV/AIDS-related pain as a chronic pain condition: implications of a biopsychosocial model for comprehensive assessment and effective management. Pain Med 2000; 1:260–73. [DOI] [PubMed] [Google Scholar]
- 3. Girach A, Julian TH, Varrassi G, Paladini A, Vadalouka A, Zis P. Quality of life in painful peripheral neuropathies: a systematic review. Pain Res Manag 2019; 2019:2091960. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Sakabumi DZ, Moore RC, Tang B, Delaney PA, Keltner JR, Ellis RJ. Chronic distal sensory polyneuropathy is a Major contributor to balance disturbances in persons living with HIV. J Acquir Immune Defic Syndr 2019; 80:568–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Erlandson KM, Allshouse AA, Jankowski CM, et al. Risk factors for falls in HIV-infected persons. J Acquir Immune Defic Syndr 2012; 61:484–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Han MM, Frizzi KE, Ellis RJ, Calcutt NA, Fields JA. Prevention of HIV-1 TAT protein-induced peripheral neuropathy and mitochondrial disruption by the antimuscarinic pirenzepine. Front Neurol 2021; 12:663373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Parast L, Cai T, Tian L. Using a surrogate marker for early testing of a treatment effect. Biometrics 2019; 75:1253–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Katz R. Biomarkers and surrogate markers: an FDA perspective. NeuroRx 2004; 1:189–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Tortelli R, Copetti M, Ruggieri M, et al. Cerebrospinal fluid neurofilament light chain levels: marker of progression to generalized amyotrophic lateral sclerosis. Eur J Neurol 2015; 22:215–8. [DOI] [PubMed] [Google Scholar]
- 10. Tortelli R, Ruggieri M, Cortese R, et al. Elevated cerebrospinal fluid neurofilament light levels in patients with amyotrophic lateral sclerosis: a possible marker of disease severity and progression. Eur J Neurol 2012; 19:1561–7. [DOI] [PubMed] [Google Scholar]
- 11. Petzold A, Hinds N, Murray NM, et al. CSF Neurofilament levels: a potential prognostic marker in Guillain-Barre syndrome. Neurology 2006; 67:1071–3. [DOI] [PubMed] [Google Scholar]
- 12. Petzold A, Brettschneider J, Jin K, et al. CSF protein biomarkers for proximal axonal damage improve prognostic accuracy in the acute phase of Guillain-Barre syndrome. Muscle Nerve 2009; 40:42–9. [DOI] [PubMed] [Google Scholar]
- 13. Mariotto S, Farinazzo A, Magliozzi R, Alberti D, Monaco S, Ferrari S. Serum and cerebrospinal neurofilament light chain levels in patients with acquired peripheral neuropathies. J Peripher Nerv Syst 2018; 23:174–7. [DOI] [PubMed] [Google Scholar]
- 14. Brierley JB. The penetration of particulate matter from the cerebrospinal fluid into the spinal ganglia, peripheral nerves, and perivascular spaces of the central nervous system. J Neurol Neurosurg Psychiatry 1950; 13:203–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Sandelius A, Zetterberg H, Blennow K, et al. Plasma neurofilament light chain concentration in the inherited peripheral neuropathies. Neurology 2018; 90:e518–e24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Esiri MM, Morris CS, Millard PR. Sensory and sympathetic ganglia in HIV-1 infection: immunocytochemical demonstration of HIV-1 viral antigens, increased MHC class II antigen expression and mild reactive inflammation. J Neurol Sci 1993; 114:178–87. [DOI] [PubMed] [Google Scholar]
- 17. Griffin JW, Wesselingh SL, Griffin DE, Glass JD, McArthur JC. Peripheral nerve disorders in HIV infection. Similarities and contrasts with CNS disorders. Res Publ Assoc Res Nerv Ment Dis 1994; 72:159–82. [PubMed] [Google Scholar]
- 18. Rance NE, McArthur JC, Cornblath DR, Landstrom DL, Griffin JW, Price DL. Gracile tract degeneration in patients with sensory neuropathy and AIDS. Neurology 1988; 38:265–71. [DOI] [PubMed] [Google Scholar]
- 19. Hahn K, Robinson B, Anderson C, et al. Differential effects of HIV infected macrophages on dorsal root ganglia neurons and axons. Exp Neurol 2008; 210:30–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Louwsma J, Brunger AF, Bijzet J, et al. Neurofilament light chain, a biomarker for polyneuropathy in systemic amyloidosis. Amyloid 2021; 28:50–5. [DOI] [PubMed] [Google Scholar]
- 21. Huehnchen P, Schinke C, Bangemann N, et al. Neurofilament proteins as a potential biomarker in chemotherapy-induced polyneuropathy. JCI Insight 2022; 7:e154395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Hayashi T, Nukui T, Piao JL, et al. Serum neurofilament light chain in chronic inflammatory demyelinating polyneuropathy. Brain Behav 2021; 11:e02084. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Heaton RK, Grant I, Butters N, et al. The HNRC 500–neuropsychology of HIV infection at different disease stages. HIV neurobehavioral research center. J Int Neuropsychol Soc 1995; 1:231–51. [DOI] [PubMed] [Google Scholar]
- 24. Antinori A, Arendt G, Becker JT, et al. Updated research nosology for HIV-associated neurocognitive disorders. Neurology 2007; 69:1789–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Heaton RK, Clifford DB, Franklin DR Jr., et al. HIV-associated neurocognitive disorders persist in the era of potent antiretroviral therapy: CHARTER Study. Neurology 2010; 75:2087–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Blackstone K, Moore DJ, Franklin DR, et al. Defining neurocognitive impairment in HIV: deficit scores versus clinical ratings. Clin Neuropsychol 2012; 26:894–908. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Lu HJ, Fu YY, Wei QQ, Zhang ZJ. Neuroinflammation in HIV-related neuropathic pain. Front Pharmacol 2021; 12:653852. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Jazebi N, Evans C, Kadaru HS, et al. HIV-related Neuropathy: pathophysiology, treatment and challenges. J Neurol Exp Neurosci 2021; 7:15–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Lanman T, Letendre S, Ma Q, Bang A, Ellis R. CNS Neurotoxicity of antiretrovirals. J Neuroimmune Pharmacol 2019; 16:130–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Evans SR, Lee AJ, Ellis RJ, et al. HIV peripheral neuropathy progression: protection with glucose-lowering drugs? J Neurovirol 2012; 18:428–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Morgello S, Estanislao L, Simpson D, et al. HIV-associated distal sensory polyneuropathy in the era of highly active antiretroviral therapy: the Manhattan HIV Brain Bank. Arch Neurol 2004; 61:546–51. [DOI] [PubMed] [Google Scholar]
- 32. Simpson DM, Kitch D, Evans SR, et al. HIV Neuropathy natural history cohort study: assessment measures and risk factors. Neurology 2006; 66:1679–87. [DOI] [PubMed] [Google Scholar]
- 33. Ellis RJ, Evans SR, Clifford DB, et al. Clinical validation of the NeuroScreen. J Neurovirol 2005; 11:503–11. [DOI] [PubMed] [Google Scholar]