Immunotherapy-Resistant Neuropathic Pain and Fatigue Predict Quality-of-Life in Contactin-Associated Protein-Like 2 Antibody Disease

Abstract

The long-term clinical outcomes and associated prognostic factors in contactin-associated protein-like 2 (CASPR2)-antibody diseases are unknown. A total of 75 participants with CASPR2 antibodies were longitudinally assessed for disability, quality-of-life, and chronic pain. Although most symptoms improved within 6 months of treatment, neuropathic pain and fatigue were the most immunotherapy refractory, and persisted for up to 6 years. Furthermore, these two factors-but not CASPR2 antibody levels or subclasses-independently predicted worse disability and quality-of-life at 24 months. Quality-of-life varied widely for any given modified Rankin Scale score, indicating a divergence between patient and clinician assessed outcomes. Further work should study the relative importance of these measures, and the immunopathogenesis underlying intractable symptoms. ANN NEUROL 2025;97:521-528.

FIGURE 1

Demographics, tumor association, and clinical features by syndrome. (A) Of 83 participants with contactin‐associated protein‐like 2 (CASPR2) antibodies, 75 were diagnosed with clinically‐related CASPR2‐antibody disease (8 had an unrelated clinical syndrome): limbic encephalitis (LE; n = 37), Morvan's Syndrome (MoS; n = 35), and peripheral nerve hyperexcitability (PNH; n = 3). (B) Of n = 75, 66 were male, and showed a median age of 66 years (range 17–82 years). Whole‐body imaging revealed 25 lesions: 12 malignant thymomas, 6 prostate cancers, 3 other cancers (small cell lung, bowel, and Merkel cell carcinoma), and 3 benign lesions. (C) Based on disease registry data, the age‐adjusted prevalence of all cancers, prostate cancer, and thymoma were overrepresented in our cohort compared with the general population. Similarly, the rate of benign prostatic hypertrophy (BPH) was overrepresented compared with the male general population. BPC = behavioral/personality change; CSF = cerebrospinal fluid; EEG = electroencephalography; EMG = electromyography; HLA = human leukocyte antigen; LE = limbic encephalitis; MoD = movement disorder; Mood = mood disturbance; MoS = Morvan's syndrome; MRI = magnetic resonance imaging; NeP = neuropathic pain; NCS = nerve conduction studies; PNH = peripheral nerve hyperexcitability. [Color figure can be viewed at www.annalsofneurology.org]

FIGURE 2

Relative symptom frequencies over time. (A) The relative frequency of symptom pairs was evaluated in all patients at the peak of disease, 6 months and 48 months. At peak, amnesia, behavioral/personality change (BPC), insomnia and dysautonomia co‐occurred most frequently, followed by neuropathic pain (NeP) and fatigue. At 6 months, NeP persisted, and fatigue became more prominent. By 48 months, as most other symptoms improved, persistent NeP and fatigue emerged as the most prominent symptom pair. (B) The frequency of co‐occurrent symptoms was evaluated in the two main syndrome groups, limbic encephalitis (LE) and Morvan's syndrome (MoS). At peak in the LE group, seizures and BPC were most frequent, whereas the MoS group showed prominent co‐occurrence of NeP, peripheral nerve hyperexcitability (PNH), dysautonomia, and insomnia. At 6 months, BPC, mood disorder, and fatigue were most prominent in the LE group, whereas in the MoS group, NeP, fatigue, and insomnia co‐occurred most frequently. By 48 months, as most other symptoms improved, NeP and fatigue were the most prominent residual symptoms in both the LE and MoS groups. MoD = movement disorder. [Color figure can be viewed at www.annalsofneurology.org]

FIGURE 3

Longitudinal outcome measures in contactin‐associated protein‐like 2 (CASPR2) antibody diseases. (A) Modified Rankin Scale (mRS) scores at the peak of disease through to 72 months. (B) EuroQol 5D Visual Analog Scale (EQ5DVAS) variation with mRS and over time (to 72 months). Deaths were carried forward such that there were 6 of 24 (25%) at 72 months. (C) Across all time points, mRS correlated with EQ5DVAS (R = −0.82, p < 2.2 × 10⁻¹⁶), and showed variation with pain severity (as measured by Chronic Pain Grading Scale [CPGS]). IT = immunotherapy; LE = limbic encephalitis; MoS = Morvan's Syndrome; mRS = modified Rankin Scale; NeP = neuropathic pain; PNH = peripheral nerve hyperexcitability; QoL = quality of life. [Color figure can be viewed at www.annalsofneurology.org]

FIGURE 4

The impact of neuropathic pain (NeP) and fatigue impact on longitudinal outcome measures and response to immunotherapy. (A) NeP and fatigue both associated with higher modified Rankin Scale (mRS) and lower EuroQol 5D Visual Analog Scale (EQ5DVAS) scores. Unadjusted (*) and adjusted (**) p values (<0.05) highlight statistically significant differences in univariate tests, and ^★ p < 0.05 and ^★★ p < 0.001 indicate statistically significant differences between curves (fitted with polynomial linear mixed effects model). (B) Symptom prevalence over time in groups divided by those treated with immunotherapy (n = 64) and only symptomatic therapies (n = 10). BPC = behavioral or personality change; CASPR2 = contactin‐associated protein‐like 2; CPGS = Chronic Pain Grading Scale; IT = immunotherapy; MoD = movement disorder; Mood = mood disturbance; NeP = neuropathic pain; PNH = peripheral nerve hyperexcitability; QoL = quality of life. [Color figure can be viewed at www.annalsofneurology.org]

FIGURE 5

Clinical and biological factors associated 24‐month outcomes and relapses. (A) Groups divided by early (<12 weeks) or late (>12 weeks) immunotherapy have Chronic Pain Grading Scale (CPGS), modified Rankin Scale (mRS), and EuroQol 5D Visual Analog Scale (EQ5DVAS) scores compared between the peak and 24 months. Multiple comparison adjusted significance observed (***Padj < 0.01). (B) Multiple logistic regression of selected clinical predictors mRS or EQ5DVAS at 24 months, or relapse at any point. A worse outcome was defined as a value above (for mRS) or below (for EQ5DVAS) the median for a given time point. Neuropathic pain (NeP) at onset (OR 15.5, 95% CI 1.4–2,173) and fatigue at onset (OR 4, 95%CI 1.1–18.5) were associated with worse mRS outcomes at 24 months and fatigue in EQ5DVAS at 24 months (OR 6.2, 95%CI 1.1–50.1), and fatigue also predicted an increased risk of relapse (OR 22.4, 95%CI 3–327.4). (C) Representative flow cytometry plots (top two panels) showing immunoglobulin G (IgG) binding (IgG mean fluorescence intensity [MFI], x‐axis) to CASPR2‐expressing human embryonic kidney cells in sera from healthy controls (left) and CASPR2‐IgG patients (right). Dotted gate quantifies CASPR2‐reactive IgGs by flow cytometry (IgG‐FACS). Lower: CASPR2‐IgG by FACS correlates well with CASPR2‐IgG levels by cell‐based assay (CBA) endpoint dilution (R = 0.90; p < 2.2 × 10⁻¹⁶) with wide variation in FACS data per endpoint dilution. (D) CASPR2‐IgG, ‐IgG1, and ‐IgG4 quantification by FACS. Those with NeP showed lower with median CASPR2‐IgG (Padj < 0.01); poor mRS and EQ5DVAS outcomes at 24 months were associated with lower median CASPR2‐IgG (Padj < 0.05), and ED5DVAS with lower CASPR2‐IgG1 (Padj < 0.05) and lower CASPR2‐IgG4 levels (Padj < 0.05). *p < 0.05; **Padj < 0.05; ***Padj < 0.01. [Color figure can be viewed at www.annalsofneurology.org]