Intensive Combination Immunotherapy and Neuroinflammation Resolution in a Child With Anti-PCA-1 (Yo) Paraneoplastic Syndrome and 2 Malignancies
Guillermo Philipps, MD1,2, Elizabeth D. Tate, MN, ARNP2, and Michael R. Pranzatelli, MD2
Abstract
Paraneoplastic cerebellar degeneration is rare and noteworthy in children. In this 7-year-old, it was documented to have occurred within a year of ataxia presentation. The instigating cancer was stage III adrenal adenocarcinoma, remitted after surgical resection at age 2. When her severe ataxia progressed, neuroinflammation was characterized by high cerebrospinal fluid Purkinje cell cytoplasmic antibody type 1 titers, oligoclonal bands, and neurofilament light chain. The immunotherapy strategy was to replace IV methyl- prednisolone, which lowered Purkinje cell cytoplasmic antibody type 1 titers without clinical improvement, with induction of adrenocorticotropic hormone/intravenous immunoglobulin/rituximab (ACTH/IVIG/rituximab) combination immunotherapy, ACTH/dexamethasone transition, and intravenous immunoglobulin maintenance. She became self-ambulatory and cerebrospinal fluid inflammatory markers regressed. Down syndrome predisposed her to a second cancer, pre-B acute lymphoblastic leukemia, 4 years later. Despite reversible cytosine arabinoside-provoked cerebellar toxicity, the ataxia is stable on monthly intravenous immunoglobulin without relapse, now 5 years after initial diagnosis. This report illustrates the use of cerebrospinal fluid biomarkers to detect, target, and monitor neuroinflammation, and successful combinations of immunotherapy to better the quality of life.
Keywords
PCA-1 syndrome, pediatric neuroinflammatory disorders, adrenocortical carcinoma, Down syndrome, OMS, pediatric paraneo- plastic cerebellar degeneration, acute lymphoblastic leukemia, cytosine arabinoside-induced ataxia, ANNA-1 (Hu) syndrome
Received May 03, 2018. Received revised June 28, 2018. Accepted for publication July 12, 2018.
Purkinje cell autoimmunity is now known to comprise a variety of demonstrable autoantibody disorders. The Purkinje cell cytoplasmic antibody type 1 (PCA-1 or anti-Yo) is an anti- onconeural autoantibody, giving rise to an ataxia- predominant paraneoplastic syndrome in women with breast or gynecologic cancers. Purkinje cell cytoplasmic antibody type 1 is prone to development of paraneoplastic cerebellar degeneration, which can be the presenting sign of cancer, or a delayed phenomenon in 30%.2,3 Studies of the adaptive and innate immune response are few. Immunotherapies, such as steroids, intravenous immunoglobulin (IVIG), and plasma exchange are used with limited success. With some excep- tions,3 the prognosis is poor; most adults become bedridden.1 We now present observations on the neuroimmunologic
profile, and clinical and immunologic responses to a novel treatment approach in a child with Purkinje cell cytoplasmic antibody type 1-induced paraneoplastic cerebellar degeneration.4
’ Department of Pediatric Neurology, Golisano Children’s Hospital of Southwest FL, Fort Myers, FL, USA
2 National Pediatric Myoclonus Center and National Pediatric Neuroinflammation Organization, Inc., Orlando, FL, USA
Corresponding Author:
Michael R. Pranzatelli, MD, National Pediatric Neuroinflammation Organization, Inc., 12001 Research Parkway, Suite 236, Orlando, FL 32826, USA.
Email: mpranzatelli@omsusa.org
CC
i BY
Case Overview
Prior to any concerns about neuroinflammatory disorders, the patient carried diagnoses of Down syndrome; stage III adreno- cortical carcinoma, which was resected at the age of 2 (along with her left kidney); static mild congenital hydrocephalus (not shunted, stable); reactive airway disease (prn albuterol); and developmental delay (speaking in sentences at 5, toilet trained at 6). The family history was pertinent for multiple sclerosis in the maternal great aunt and gene-loaded for cancer on the paternal side (leukemia) father deceased from grade IV glio- blastoma multiforme-and maternal side (breast cancer). At the age 7 years, she presented with progressive ataxia, which was not recognized as being paraneoplastic until 3 months later. Four years afterward, she developed acute lymphoblastic leukemia, as predisposed by trisomy 21. A p53 mutation, caus- ing risk of recurrent malignancies in keeping with Li Fraumeni syndrome, was discovered. It is noteworthy that the appearance of the paraneoplastic syndrome was delayed 5 years, and the interval between the 2 malignancies spanned 9 years. Addi- tional clinical description and a detailed clinical course sum- mary are provided in Table 1.
Methods
Clinical
The patient was referred to the National Pediatric Myoclonus Center for second evaluation of clinical deterioration. Parents gave written informed consent for their child to participate in this institutional review board-approved study of immunological abnormalities in opsoclonus-myoclonus as a related neuroinflammatory/paraneoplas- tic disorder (SIU SOM, Springfield, Illinois). Clinical data were collected, and extra cerebrospinal fluid and blood for research pur- poses were obtained from lumbar puncture and venipuncture per- formed for clinical reasons. Immunotherapy was given in clinical practice, not a drug trial, by the local treating physicians. Given the clinical gravity of paraneoplastic cerebellar degeneration and the persistent ataxia, parents were reconsented for a second lumbar puncture to rule out ongoing neuroinflammation.5 Years later, oncol- ogists at the referring hospital performed 2 lumbar punctures to rule out central nervous system leukemic infiltration and subsequent laboratory testing. Western institutional review board (Puyallup, Washington) conferred exempt review status for retrospective data analysis. A modified Opsoclonus Myoclonus Evaluation Scale (nys- tagmus substituted for opsoclonus) was used to compute Total Score (0-36) from videotapes.6 Retrospective gait ataxia scoring by the first author was used to summarize ataxia evaluations from the clinic. The patient’s mother gave written approval for publication of this report with videotapes, including recognizable facial images neces- sitated by the presence of nystagmus.
Laboratory
Lymphocyte subsets were measured in cerebrospinal fluid and blood by flow cytometry in the clinical laboratory as previously described.6 Lymphocytes gated were 1633 from 14 mL of fresh cerebrospinal fluid from the first visit lumbar puncture and 6838 from 14.5 mL for the second lumbar puncture at the National Pediatric Myoclonus Cen- ter. Cerebrospinal fluid oligoclonal bands not found in serum (positive
if ≥2) were measured by isoelectric focusing with immunofixation at ARUP Lab (Salt Lake City, Utah).
Cerebrospinal fluid and serum immunobiomarkers were measured in Dr Pranzatelli’s neuroimmunology laboratory by enzyme-linked immunosorbent assay using commercial kits, as per the manufactur- er’s instructions. The neurofilament light chain kit was purchased from Ulman Diagnostics (Umeá, Sweden; lot no. 70189); the micro- glial/macrophage marker soluble chitinase 3-like 1 (sCHI3L1) kit, from R&D systems, Inc (Minneapolis, Minnesota; lot no. 298223). Both were used as previously described.7,8
A commercial paraneoplastic serologic evaluation was performed by Mayo Clinic Lab (Rochester, Minnesota) for ANNA Type 1, 2, and 3, anti-glial nuclear antibody type 1, Purkinje cell cytoplasmic anti- body (Purkinje cell cytoplasmic antibody type 1, 2, and Tr), amphi- physin antibody, and CRMP-5-immunoglobulin G; antibodies to GAD65 and to N-methyl-D-aspartate (NMDA), y-aminobutyric acid (GABA), or a-amino-3-hydroxy-5-methyl-4-isoxazoleproprionic acid (AMPA) receptors; SRP, striational, P/Q-Type calcium channel, N- Type calcium channel, ACh receptor binding, AChR ganglionic neu- ronal, and neuronal (V-G) K+ channel antibodies.
Published control data from Dr Pranzatelli’s neuroimmunology laboratory were obtained from children with noninflammatory neu- rological disorders and non-neurological disorders who underwent lumbar puncture as part of clinical diagnostic testing. Those data provided medians and reference ranges for use in comparison with the index patient.
Results
Clinical and Radiological Response to Immunotherapy
The treatment sequence, involving a multiplicity of immu- notherapeutic agents, is shown in Figure 1A. In brief, the initial regimen of methylprednisolone and prednisone after only a single treatment of intravenous immunoglobulin was changed substantially for the lack of clinical effectiveness. An ongoing regimen of intravenous immunoglobulin was supplemented by a course of rituximab followed by initiation of adrenocortico- tropic hormone (ACTH) after IV dexamethasone, then changed to oral dexamethasone after ACTH weaning and discontinua- tion. Maintenance intravenous immunoglobulin and oral dex- amethasone were extended during the cancer chemotherapy for acute lymphocytic leukemia.
The cumulative sequence of diagnoses is depicted in Figure 1B. Besides the initial diagnoses of ataxia, paraneoplastic cer- ebellar degeneration, Purkinje cell cytoplasmic antibody type 1, and autoimmune thyroiditis, the patient was subsequently shown to have vitamin D insufficiency and received vitamin D supplementation. At 46 to 50 months before the presentation of acute lymphocytic leukemia, the patient began to experience mild ataxia regression 1 week before intravenous immunoglo- bulin was due (ie, 3 weeks after dose) on several occasions, which remitted quickly after intravenous immunoglobulin was given. After exposure to cytosine arabinoside, the patient’s ataxia worsened-a known cerebellar toxicity of cytosine ara- binoside-but it recovered. Autoimmune thyroiditis was diag- nosed on the basis of positive thyroglobulin at 415 IU/mL (normal < 116 IU/mL) and elevated thyroperoxidase antibodies
| Clinic Visit | Time After Ataxia (Months) | Drug/Biological Treatments | History and Neurological Examination |
|---|---|---|---|
| 1 | 1 | None | New-onset ataxia, progressing. Intermittent horizontal nystagmus, low tone. Not cooperative for proprioception testing. Gait wide based and unsteady. Falls over easily with changes in position. One month prior MRI brain report: enlarged lateral ventricles, unchanged from prior studies. Dx: ataxia of unknown etiology |
| 2 | 2 | IVIG | g/kg/d × 2 days | Worsening ataxia needs to hold onto walls. CSF: normal protein/WBC. Normal MRI c/t/l-spine. EMG/NCV normal. Horizontal nystagmus, DTR reduced. Very unsteady gait |
| 3 | 4 | IVIG I g/kg/d × 2 days, IV MPRED 30 mg/kg/d × 3 days, ☒ PRED taper (Img/kg/d × 3 days, 0.75 mg/kg/d × 3 days, ☒ 0.5 mg/kg/d × 3 days and stopped) ☒ | Ataxia persists. Using a walker. No significant improvement. Exam unchanged. Paraneoplastic panel positive for anti-yo (PCA-1) antibodies: serum 1:15360, CSF 1:256. Repeat CSF 1WBC, protein 38.6 |
| 4 | 5 | IVIG I g/kg/day x I day (repeated monthly), IV MPRED ☒ ☒ 30 mg/kg/d x I day (repeated monthly) | Ataxia not clearly progressing anymore. Can walk with hand held. Difficulty using a walker. Repeat PCA-1 titers: Serum I:1920, CSF I:128. P53 mutation detected-missense mutation in exon 8 |
| 5 | 6 | Monthly IVIG I g/kg/d x I day, and IV MPRED 30 mg/kg/d x ☒ 1 day | Ataxia again progressing. CSF with oligoclonal bands. Exam: highly unsteady with standing. Severe ataxia |
| 6 | 8 | Monthly IVIG Ig/kg/d x I day, PO DEX 7 mg/m2/d x 3 days, ☒ ☒ RTX 300 mg/m2 × 4 weekly ☒ | Ataxia stabilized and improved. Falling less often. Can now throw a ball from standing. On exam, gait wide based, but can bend over, but falls |
| 7 | 10 | Monthly IVIG Ig/kg/d x I day. ACTH: 75 IU/m2 BID x ☒ ☒ I week, 75 IU/m2 daily x 4 weeks, 75 IU/m2 QOD x ☒ 3 weeks, 65 IU/m2 QOD x I week, 55 IU/m2 QOD x ☒ ☒ 1 week | Ataxia worsened again 2 months after rituximab completed. Initiated ACTH 4 mo treatment course with improvements. Minor side effects of increased BP. Exam: sways with standing position. Base in gait narrower. Turns and bends over without falling. Repeat MRI brain- cerebellar atrophy |
| 8 | 12 | Monthly IVIG I g/kg/d x I day, ACTH: 45 IU/m2 QOD > ☒ ☒ I week, 35 IU/m2 QOD x I week, 25 IU/m2 QOD x ☒ I week, 15 IU/m2 QOD x I week, 10 IU/m2 QOD x ☒ I week, 5 IU/m2 QOD x I week | Seizure and altered mental status secondary to elevated BP. Started on levetiracetam. BP managed with enalapril. Balance improving. Can pick up things without falling. Climbing up and down stairs. Falls with quick turns, but less often. On exam, cushingoid with weight gain. Mild irritability |
| 9 | 15 | Monthly IVIG I g/kg/d x I day, PO DEX 7 mg/m2/d x 3 days/ months | CSF PCA-I titer: 1:8. Off ACTH for past month. Balance continued to improve. Can run. Intermittent worsening of balance when ill. Off levetiracetam. BP normalized |
| 10 11 | 22 37 | Monthly IVIG I g/kg/d x I day, PO DEX 7 mg/m2/d x 3 days/ ☒ ☒ months Monthly IVIG I g/kg/day x | day, PO DEX 7 mg/m2/d x ☒ 2 days/months | Balance stable. Does not sleep well despite melatonin. Per PT report: Can walk 200 to 300 feet without falling (compared to 10 feet, 12 months prior). Mild-moderate unsteadiness with standing on 2 feet. Can stand on one leg with minimal assistance. Stance 10 to 15 inches (was >18) Gait stable to improved. Still with intermittent worsening of ataxia when ill. Continued sleeping problems, failed clonidine due to low BP. Exam, stands steadily with minimally wide based stance. Minimal unsteadiness |
| 12 13 | 48 51 | Monthly IVIG I g/kg/d x I day; PO; DEX 7 mg/m2/d x ☒ ☒ 2 days/months. IV MPRED 2 mg/kg/d × 2 days; IV DEX 7 mg/m2 × | day ☒ Monthly IVIG I g/kg/d x I day, COG AALL 11311 ☒ | Ataxia stable. However, ambulation limited by left hip/knee pain for past 3 months. Left hip effusion found. Suspicion for autoimmune process. Improved with IV steroids New onset thrombocytopenia during steroid wean. Elevated inflammatory markers. Bone marrow positive for Pre-B ALL. CSF negative for malignancy. ALL in Remission at day 15. Cytogenetics showed near tetraploidy. Heterozygous for TPMT. Repeat PCA-1: serum undetectable, CSF 1:32 |
(continued)
| Clinic Visit | Ataxia (Months) | Drug/Biological Treatments | History and Neurological Examination |
|---|---|---|---|
| 14 | 57 | Monthly IVIG I g/kg/d x I day, chemotherapy | At home, ataxia stable. Developed Altered mental status and worsened ataxia secondary to Ara-c toxicity. ALL in remission. MRI brain unchanged from 2013 study with cerebellar atrophy and ventriculomegaly |
| 15 | 62 | Monthly IVIG I g/kg/d x I day, maintenance chemotherapy | Recovered from Ara-c toxicity. Mental status back to her normal. Cognition also improving. Ataxia stable. Continues to be able to do stairs. ALL in remission. Exam: child happy and interactive. Fairly steady gait, mild wide base |
| 16 | 68 | Monthly IVIG I g/kg/d x I day, maintenance chemotherapy | Continues in remission from ALL. Ataxia remains unchanged. Repeat CSF PCA-1 antibody titer 1:8. Exam unchanged from last visit |
Abbreviations: ACTH, adrenocorticotropic hormone; ALL, acute lymphoblastic leukemia; Ara-c, cytosine arabinoside; BID, twice a day; BP, blood pressure; COG, Children’s Oncology Group; DEX, dexamethasone; CSF, cerebrospinal fluid; IVIG, intravenous immunoglobulin; MPRED, methylprednisolone; PO, orally; PRED, prednisone; RTX, rituximab; TPMT, thiopurine S-methyltransferase gene; QOD, once a day.
ªCOG AALL 1131: Chemotherapy days 1-14: Ara-C IT 70 mg/m2, Vincristine IV 1.5 mg/m2/dose days | and 8, Methotrexate 15 mg IT day 8, Prednisone 30 mg/m2/ dose BID PO days 1-14, PEG-Asparaginase IV 2500 units/m2 day 4. Days 15-29: Vincristine IV 1.5 mg/m2/dose day 15 and day 22, Prednisone 30 mg/m2/dose BID PO days 15-29, Methotrexate 15 mg IT day 29.
at 37.9 IU/mL (normal < 9.0 IU/mL). The thyroperoxidase antibodies concentration fell on immunotherapy (Figure 1H). She remains on levothyroxine therapy.
The ataxia response is depicted in Figure 1I. By the end of multimodal induction, ataxia severity had declined by 50%, with the functional improvement of beginning to walk inde- pendently, though still ataxic. Further improvement occurred despite switching ACTH to dexamethasone. On long-term intravenous immunoglobulin maintenance therapy, ataxia remained at a low level before and after a small rise associated with acute lymphocytic leukemia/cytosine arabinoside.
Using the modified Opsoclonus Myoclonus Evaluation Scale (Figure 1J), Total Score declined by 24% on ACTH/ intravenous immunoglobulin/rituximab induction therapy (Supplemental materials, compare Video 1 and Video 2), although remaining in same moderate severity category (13- 24 points). Comparison of the neurologic examinations is shown in Table 2. On long-term maintenance therapy, how- ever, Total Score fell into the upper end of the mild range (0- 12 points), a 48% drop from the initial score (Supplemental Material, Video 3). As the Opsoclonus Myoclonus Evaluation Scale also evaluates development, Total Score might not be expected to drop much further due to Down syndrome.
Neuroimaging of the brain (Figure 2) revealed correlates. The magnetic resonance imaging (MRI) that was obtained 2 months prior to Purkinje cell cytoplasmic antibody type 1 diagnosis (about 2 weeks into the ataxia symptoms) did not show paraneoplastic cerebellar degeneration (Figure 2A and D). Presumably, the cerebellum was normal because it was early in the disease course. How fast the degeneration devel- oped thereafter is unknown, except to say that it had occurred by the next MRI 10 months after the initial scan (Figure 2B and E). The radiologic appearance of paraneoplastic cerebellar
degeneration did not improve on immunotherapy nor did it progress (Figure 2C and F).
Neuroimmunologic Studies
At the first diagnostic evaluation, the main neuroimmunologic observations were mildly increased the frequency of cerebrosp- inal fluid B cells (1.6%), positive oligoclonal bands (6), and elevated Purkinje cell cytoplasmic antibody type 1 titer (Figure 1 C-E). Cerebrospinal fluid immunoglobulin G index, immu- noglobulin G synthesis rate, immunoglobulin G/albumin ratio, protein, and glucose were normal. There was no cerebrospinal fluid pleocytosis (WBC 1/cu mm; RBC 0). The cerebrospinal fluid leukocyte differential was 80% lymphocytes and 20% monocytes. The cerebrospinal fluid CD4/CD8 T-cell ratio was low at 1.8 (normal 2.5-3). The neuronal/axonal marker neuro- filament light chain was elevated 10-fold in cerebrospinal fluid at 3254 ng/ml (controls, 300 ng/ml) and also in serum at 722 ng/ml (controls, 22 ng/mL), and the neurofilament light chain cerebrospinal fluid: serum ratio was 4.5. The serum Purkinje cell cytoplasmic antibody type 1 titer was grossly elevated at 1:15360 (Figure 1F). The blood B cell frequency of 9.8% was not elevated (Figure 1G).
At 7 months on intensified combination immunotherapy, there was amelioration of adaptive immunity: undetectable cerebrospinal fluid B cells, and reduction in oligoclonal bands count (3) and Purkinje cell cytoplasmic antibody type 1 titers (1:8). WBC count was 4/cu mm (81% lymphocytes/19% mono- cytes); RBC 0. Cerebrospinal fluid immunoglobulin G index was normal; immunoglobulin G index, slightly elevated at 0.71. The cerebrospinal fluid CD4/CD8 T-cell ratio normal- ized. The frequency of yo T-cells, primarily considered to be innate immune cells, was normal at 2.6% in cerebrospinal fluid
A
B
C
Pred-3
Ara-c Rec.
2.0-
Pred-2
Ara-c Atax.
Chemo
ALL
% CSF B Cells
1.5
Dex-PO
ACTH
Regres.
RTX
Vit D Insuf.
1.0
Dex-IV
IVIG-2
Thyroiditis
Pred-1
PCA-1
Mpred
PCD
0.5
IVIG-1
Ataxia
0.0
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
Treatment Time Course (months)
Treatment Time Course (months)
D
E
F
8-
300
Serum PCA-1 Titer x 1000
20
CSF OCB No.
CSF PCA-1 Titer
6
15
200
4
10
100
2
5
0
0
0
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
G
H
10-
Serum [TPO Ab] (IU/ml)
40
% Blood B Cells
8
30
6
20
4
2
10-
0
0
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
J
8
36-
30
Gait Ataxia
6
Total Score
24
4
18
12
2
6
0
0
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
Treatment Time Course (months)
Treatment Time Course (months)
| Feature | Visit 1 | Visit 2 |
|---|---|---|
| Cooperative | Mostly | Mostly |
| Follows commands | With coaching | With coaching |
| Dysarthria | Moderate | Mild |
| Sentences | No | 3 words |
| Cranial nerves | Intact | Intact |
| DTR | Absent | Absent |
| Ankle clonus | No | No |
| Extensor plantar | No | No |
| Block stacking | 0 of 8 (right hand) | 3 of 8 (right) |
| 3 of 8 (left) | 6 of 8 (left) | |
| Paperclip in bottle | With anchoring | With anchoring |
| Finger-nose dysmetria | Moderate | Mild |
| Muscle tone | Decreased | Decreased |
| Muscle strength | Normal | Normal |
| Standingª | Wide base, backstepping | Wide base |
| Gait | Wide base, ataxic | Wide base, ataxic |
| One-foot balance | No | 1 second |
| Hopping | No | No |
| Falling | Several times | A few times |
| Able to run | No | No |
| Ball throwing | 0 of 3 tries | 1 of 3 times |
| Ball catching | 0 of 3 tries | 1 of 3 tries |
| Getting off floor | 4-limb push off | 4-limb push off |
Abbreviations: NPMC, National Pediatric Myoclonus Center.
aOne-foot gap between feet while standing on visit I.
and 10.8% in blood. The concentration of the M2 macrophage marker CHI3L1 in serum was 36 ng/ml at the low range of controls (431 ng/ml). Blood B cell frequency was 0%.
When acute lymphocytic leukemia presented (on predni- sone 15 mg/d and intravenous immunoglobulin 8 days prior), cerebrospinal fluid oligoclonal bands was 2, immunoglobulin G synthesis rate 11.5 mg/d (vs < 8), cerebrospinal fluid immunoglobulin G 8.3 mg/dL (vs ≤ 6), and albumin index (10.9 vs ≤ 9). These were interpreted as mild elevations.
Discussion
This case introduces several novelties. First, intensified com- bination immunotherapy induced marked clinical improve- ment, whereas prior methylprednisolone/intravenous immunoglobulin also reduced Purkinje cell cytoplasmic anti- body type 1 titers but with little clinical benefit. Second, signs of neuroinflammation steadily declined, all but a few normal- izing, corresponding to functional improvement. Third, no major relapses occurred on dexamethasone/intravenous immu- noglobulin maintenance therapy, despite Down syndrome- associated leukemia and chemotherapy. Fourth, the features are consistent with adult-onset Purkinje cell cytoplasmic anti- body type 1, but with better outcome. Fifth, cerebrospinal fluid neurofilament light chain was elevated, as in opsoclonus-myo- clonus syndrome and anti-ANNA-1 (Hu) syndrome, suggesting a shared immunopathologic aspect of otherwise distinct clin- ical syndromes. Sixth, the patient had concomitant
autoimmune thyroiditis, which responded to the same immu- notherapy given for the paraneoplastic syndrome. Seventh, the patient was vitamin D insufficient, a common association with autoimmune disorders. Eight, there were two malignancies: the first, rare and previously not known to be associated with para- neoplastic cerebellar degeneration, the second, associated with Down syndrome.
The “Time is Brain” motto has been applied to the cerebel- lum: “Time is cerebellum.”9 Despite the cerebellum’s capacity for compensating and restoring lost functions, the therapeutic opportunity for intervention occurs early in patients with cer- ebellar diseases, particularly immune ataxias. Advanced cell loss degrades cerebellar “reserve,” hastening the transition from a restorable or treatable state to an untreatable one.9 Timely immunotherapy is necessary to treat neuroinflamma- tion comprised of the adaptive immune response of B cells and T cells, among others, and possible involvement of the innate immune system, which is involved in neurodegeneration. Yo- paraneoplastic cerebellar degeneration tumors are infiltrated by large numbers of B and T cells, some organized in tertiary lymphoid structures, and Yo-paraneoplastic cerebellar degen- eration-manifesting ovarian carcinomas harbor at least 1 genetic alteration of Yo-antigens thought to trigger the break- down of immune tolerance.1º There is a differential genetic susceptibility to anti-Yo per cancer with primary human lym- phocyte antigens class II involvement.
The combination of ACTH, intravenous immunoglobulin, and rituximab, with or without transitioning to dexamethasone for maintenance prior to weaning, has been successful in children with opsoclonus-myoclonus syndrome, both clini- cally and against neuroinflammation.6 Rituximab (anti- CD20) targets B cells. In a pilot study of rituximab in 9 adult patients (anti-Yo and anti-Hu), 3 patients improved ≥ 1 Ran- kins Scale point after monthly IV rituximab 375 mg/m2.12 Com- bination immunotherapy with dexamethasone, intravenous immunoglobulin, and rituximab (DEXIR-CI) also has similar effects.13 Rituximab influences neuroinflammation whether or not the clinical benefit occurs during the 4-week infusions or thereafter. In the broader view, there are now several different treatment protocols for paraneoplastic disorders available to child neurologists13-16 so clinical deterioration, partial response, and relapse can be addressed by forward-thinking, biomarker-assisted initial treatment planning or mid-course corrections.
In comparison of our case with the Purkinje cell cytoplasmic antibody type 1 syndrome in adults, early cerebrospinal fluid studies in adults reveal lymphocytic pleocytosis, oligoclonal bands, and elevated protein, but our patient did not have pleo- cytosis or elevated protein. The B-cell frequency was lower than usually found in OMS. As with our case, anti-Purkinje cell cytoplasmic antibody type 1 antibodies may persist for years.2 Serum Purkinje cell cytoplasmic antibody type 1 has recently been reported outside the context of cancer in 77% of children with attention deficit hyperactivity disorder (ADHD) and 22% of controls in association with elevated interleukin 6 (IL-6) and IL-10 serum concentrations.17 Pending replication
A
B
C
A
A
P
cm
P
1 cm
D
E
F
H
of that finding by other investigators, the clinical significance, and presence of any such parallels in adults remains uncertain.
The striking elevation of cerebrospinal fluid neurofilament light chain concentration in pediatric-onset paraneoplastic neu- rological disorders is significant. An elevated cerebrospinal fluid to serum neurofilament light chain ratio indicates intrathecal concentration. Now described in pediatric-onset ANNA-1 (Hu) paraneoplastic syndrome18 and OMS,7 it is a useful biochemical measure. Well known to have utility in assessing neuroinflammation in multiple sclerosis and related disorders in adults, the opportunity to utilize it presents itself to child neurologists.
Cytarabine-induced ataxia due to toxicity, which degrades cytoskeleton components like neurofilament,19 has been well described.2º The antineoplastic and immunosuppressant drug
acts through inhibition of DNA polymerase. With intrathecal administration, cerebrospinal fluid cytarabine levels decline with a half-life of 2 hours. The incidence of cerebellar ataxia is up to 14%.21 In a study of 418 patients aged 2 to 74 years with leukemia or lymphoma, 8% developed severe cytarabine- induced cerebellar toxicity, especially if > 50 years old, regard- less of gender, diagnosis, or regimen.22 Drug dose and schedule, cumulative dose, renal and hepatic dysfunction, and use of neurotropic antiemetic drugs can also affect risk.21 In our patient, it is possible that paraneoplastic cerebellar degenera- tion may have made the patient more vulnerable to the cere- bellar toxicity, which was clinically reversible; however, the cerebellum, with its extended postnatal development, is partic- ularly sensitive to toxic agents, even in patients without para- neoplastic cerebellar degeneration.
Cerebellar abnormalities associated with Down syndrome and a Down syndrome murine model includes reduced cere- bellar volume and granule cells.23 Although they may contrib- ute to the hypotonia of Down syndrome, as exhibited by this patient, they do not account for her substantial cerebellar/ver- mian atrophy or the relatively short course of its development. Pre-paraneoplastic cerebellar degeneration/Purkinje cell cyto- plasmic antibody type 1 neuroimaging did not display cerebel- lar hypoplasia or appreciable atrophy.
Whereas only 0.2% of all pediatric malignant cancers are adrenocortical tumors,24 children with Down syndrome are well known to have a greatly increased risk of leukemias.25 For acute lymphocytic leukemia, their event-free and overall survival is poorer than in non-Down syndrome acute lympho- cytic leukemia.25 Although the 5-year event-free survival for adrenocortical tumors is only 54%,24 our patient did well. The outcome is better with localized tumors of small volume.26 The patient thus far is responding well to chemotherapy. Paraneo- plastic cerebellar degeneration also occurs rarely with other pediatric-onset cancers, such as Hodgkin disease.27
This patient’s autoimmune thyroiditis, found on screening, presented differently than acute “steroid-responsive encephalo- pathy associated with autoimmune thyroiditis” (alias Hashimoto encephalopathy),28,29 with its requirement for negative brain neu- roimaging and exclusion of all other causes. Here, the autoim- mune thyroiditis most likely reflects the increased risk of Hashimoto thyroiditis and Grave disease in Down syndrome,3 and of having subsequent autoimmune disorders in addition to the initial one in children with perturbed autoimmunity. The thyroi- ditis was responsive to immunotherapy and has not returned.
Vitamin D insufficiency or deficiency is more common in patients with autoimmune disorders and has been associated with multiple sclerosis, systemic lupus erythematosus, poly- myositis and dermatomyositis, rheumatoid arthritis, Behçet disease, type 1 diabetes mellitus, and systemic scleroderma.31 Besides autoimmune disorders, the inestimable consequences of vitamin deficiency, which is now a pandemic, include can- cers, infectious disease, cardiovascular diseases, and childhood dental caries and periodontitis.32 Testing for 25-OH-vitamin D levels and supplementing with vitamin D if low is highly rec- ommended in patients with autoimmune disorders.31
Conclusion
The clinical importance of our observations is that intensive, long-term, biomarker-based, combination immunotherapy made a difference in this rare child with paraneoplastic cere- bellar degeneration and anti-Purkinje cell cytoplasmic antibody type 1 paraneoplastic syndrome, both in quality of life and amelioration of neuroinflammation. Given the better outcome than would be predicted from poor treatment responses in adults, this immunotherapy approach should be pursued in oth- ers who share the diagnosis. Purkinje cell cytoplasmic antibody type 1 antibodies should be measured in children with ataxia or paraneoplastic cerebellar degeneration presentation, even if years after cancer treatment. Alertness to the increased risk
of a second autoimmune disorder and/or malignancy and test- ing for vitamin D deficiency/insufficiency should be routine. Such patients need long-term follow-up to stay the course.
Acknowledgments
The authors thank Nathan R. McGee, B.S., for performing the neuro- filament light chain and sCHI3L assays at the National Pediatric Myoclonus Center, the patient, and her family. They also thank the following individuals at Golisano Children’s Hospital of SWF: Michael Weiss, MD, Florida Radiology Consultants, and the entire hematology/oncology team: Craig MacArthur MD, Cameron Nichol- son MD, Emad Salman MD, Kelly Sawczyn MD, Pamela Bolton ARNP, Jennifer Vega ARNP, the medical assistants, nurses, customer service representatives who made getting approval of the patient’s treatment course possible, and former attending Susan Alisanski MD.
Author Contributions
GP treated the patient, gathered the clinical and laboratory information, wrote the case description and made the clinical table, and selected the MRI scan images. ET performed the lumbar punctures at the National Pediatric Myoclonus Center, gathered the histories, examined the patient, scored the videotapes, and made the movies from them. MP examined the patient, conceptualized the manuscript, interpreted the data and flow cytometry reports, graphed the data, and drafted the manuscript. All authors read, revised, and approved the final manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Supplemental Material
Supplemental material for this article is available online.
Ethical Approval
This study was approved by the Institutional Review Board of SIU SOM (Springfield, IL). Written informed consent for the collection of cerebrospinal fluid and peripheral blood samples at the NPMC was obtained from the parent, giving consent for herself and on behalf of her cognitively impaired, minor child. Western IRB (Puyallup, WA) granted permission for retrospective analysis of all the data reported.
References
1. Venktraman A, Opal P. Paraneoplastic cerebellar degeneration with anti-Yo antibodies - a review. Ann Clin Transl Neurol. 2016;3(8):655-663.
2. Finstere J, Unterberger U, Grisol W. Asymptomatic persistence of anti-Yo antibodies for 5 years with relapse of malignancy. Neu- ropathology. 2007;27(3):300-304.
3. Cui D, Xu L, Li WY, Qian WD. Anti-Yo positive and late-onset paraneoplastic cerebellar degeneration associated with ovarian carcinoma. A case report. Medicine. 2017;96(32):e7362. doi: 10.1097/MD.0000000000007362.
4. Philipps G, Alisanski SB, Pranzatelli M, Clardy SL, Lennin VA, Mckeon A. Purkinje cell cytoplasmic antibody type 1 (anti-Yo)
autoimmunity in a child with Down Syndrome. JAMA Neurology. 2014;71(3):347-349.
5. Bielekova B, Pranzatelli MR. Promise, progress, and pitfalls in the search for CNS biomarkers in neuroimmunological diseases: a role for CSF immunophenotyping. Sem Pediatr Neurol. 2017; 24(3):229-239.
6. Pranzatelli MR, Tate ED, Travelstead AL, Colliver JA. Long- term cerebrospinal fluid and blood lymphocyte dynamics after rituximab for pediatric opsoclonus-myoclonus. J Clin Immunol. 2010;30(1):106-113.
7. Pranzatelli MR, Tate ED, McGee NR, Verhulst SJ. CSF neurofi- lament light chain is increased in OMS, decreasing with immu- notherapy, and other pediatric neuroinflammatory disorders. J Neuroimmunol. 2014:266(1-2):75-81.
8. Pranzatelli MR, Tate ED, McGee NR. Microglial/macrophage markers CHI3L1, sCD14, and sCD163 in CSF and serum of pediatric inflammatory and non-inflammatory neurological dis- orders: a case-control study and reference ranges. J Neurol Sci. 2017:381:285-290. doi: 10.1016/j.jns.2017.09.006.
9. Mitoma H, Manto M, Hampe CS. Time is cerebellum. Cerebel- lum. 2018;17(4):387-391. doi: 10.1007/s12311-018-0925-6.
10. Small M, Treilleux I, Couillault C, et al. Genetic alterations and tumor immune attack in Yo paraneoplastic cerebellar degenera- tion. Acta Neuropathol. 2018;135(4):569-579. doi: 10.1007/ s00401-017-1802-y.
11. Hilary RP, Ollila HM, Lin L, et al. Complex HLA association in paraneoplastic cerebellar ataxia with anti-Yo antibodies. J Neu- roimmunol. 2018;315:28-32.
12. Shams’ili S, de Beukelaar J, Gratama JW, et al. An uncontrolled trial of rituximab for antibody associated paraneoplastic neurolo- gical syndromes. J Neurol. 2006;253(1):16-20.
13. Pranzatelli MR, Tate ED. Dexamethasone, intravenous immuno- globulin, and rituximab combination immunotherapy for pediatric opsoclonus-myoclonus syndrome. Pediatr Neurol. 2017;73: 48-56.
14. Pranzatelli MR, Tate ED, Allison TJ. 6-Mercaptopurine modifies cerebrospinal fluid T-cell abnormalities in pediatric opsoclonus- myoclonus as steroid sparer. Clin Exp Immunol. 2017;190(2): 217-225. doi: 10.1111/cei.13015.
15. Pranzatelli MR, Tate ED, Alber M, et al. Rituximab, IVIG, and tetracosactide (ACTH 1-24) combination immunotherapy (“RITE-CI”) for pediatric opsoclonus-myoclonus syndrome: immunomarkers and clinical observations. Neuropediatrics 2017. doi: 101055/s-0037-1609038.
16. Pranzatelli MR, Allison TJ, Tate ED. Effect of low-dose cyclo- phosphamide, ACTH, and IVIg combination immunotherapy on neuroinflammation in pediatric-onset OMS: a retrospective pilot study. Eur J Paediatr Neurol. 2018;22(4):586-594. doi: 10.1016/ j.ejpn.2018.02.009.
17. Donfrancesco R, Nativio P, Di Benedetto A, et al. Anti-Yo anti- bodies in children with ADHD: First results about serum cyto- kines. J Atten Disord. 2016: pii: 10870547116643387.
18. Pranzatelli MR, McGee NR. Neuroimmunology of OMS and ANNA-1/anti-Hu paraneoplastic syndromes in a child with neu- roblastoma. Neurol Neuroimmunol Neuroinflamm. 2017;5(2); e433; doi 10.1212/NXI.0000000000000433
19. Koros C, Kitraki E. Neurofilament isoform alterations in the rat cerebellum following cytosine arabinoside administration. Toxi- col Lett. 2009;1898(3):215-218.
20. Sylvester RK, Fisher AJ, Lobell M. Cytarabine-induced cerebel- lar syndrome: case report and literature review. Drug Intell Clin Pharm. 1987;21(2):177-180.
21. Baker WJ, Royer GL, Weiss RB. Cytarabine and neurologic toxi- city. J Clin Oncol. 1991;9(4):679-693.
22. Herzig RH, Hines JD, Herzig GP, et al. Cerebellar toxicity with high-dose cytosine arabinoside. J Clin Oncol. 1987; 5(6):927-932.
23. Moldrich RX, Dauphinot L, Laffaire J, Rossier J, Potier MC. Down syndrome gene dosage imbalance on cerebellum develop- ment. Prog Neurobiol. 2007;82(2):87-94.
24. Xu X, Serrgi C. Pediatric adrenal cortical carcinomas: histopatho- logical criteria and clinical trials. A systematic review. Contemp Clin Trials. 2016;50:37-44.
25. Maloney KW, Taub JW, Ravindranath Y, Roberts I, Vyas P. Down syndrome preleukemia and leukemia. Pediatr Clin North Am. 2015;62(1):121-137.
26. Cecchetto G, Ganarin A, Bien E, et al. Outcome and prognostic factors in high-risk childhood adrenocortical carcinomas: a report from the European Cooperative Study Group on Pediatric Rare Tumors (EXPERT). Pediatr Blood Cancer. 2017;64(6). doi: 10.1002/pbc.26368.
27. Hahn A, Claviez A, Brinkmann G, Altermatt HJ, Schneppen- heim R, Stephani U. Paraneoplastic cerebellar degeneration in pediatric Hodgkin disease. Neuropediatrics. 2000;31(1): 42-44.
28. Hilberath JM, SAchmidt H, Wolf GK. Steroid-responsive ence- phalopathy associated with autoimmune thyroiditis (SREAT): case report of reversible coma and status epilepticus in an ado- lescent patient and review of the literature. Eur J Pediatr. 2014; 173(10):1263-1273.
29. Olmez I, Moses H, Sriram S, Kirschner H, Lagrange AH, Pawate S. Diagnostic and therapeutic aspects of Hashimoto’s encephalo- pathy. J Neuro Sci. 2013;331(1-2):67-71.
30. Aversa T, Crisafulli G, Zirilli G, De Luca F, Gallizzi R, Valenzise M. Epidemiological and clinical aspects of autoimmune thyroid diseases in children with Down’s syndrome. Ital J Pediatr. 2018; 44(1):39. doi: 10.1186/s13052-018-0478-9.
31. Pelajo CF, Lopez-Benitez JM, Miller LC. Vitamin D and auto- immune rheumatologic disorders. Autoimmun Rev. 2010;9(7): 507-510.
32. Holick MF. The vitamin D deficiency pandemic: approaches for diagnosis, treatment and prevention. Rev Endocr Metab Disord. 2017;18(2):153-165.