Measles Vaccine Strain Genotype A from the Skin Rash of a DiGeorge Patient on a TNF Inhibitor

Abstract

Isolation of measles virus is typically from respiratory, blood or urine specimens. We describe the first known case of measles vaccine-associated disease in a patient on TNF inhibitor therapy in which genotype A Edmonston vaccine strain virus was identified from skin scrapings of the patient’s rash.

Introduction

Due to the marked decreased incidence of measles in developed countries after introduction of measles vaccination, it has become increasingly difficult to distinguish the measles rash from other diseases and serology has become a widely used diagnostic marker [1]. PCR-based testing of throat and urine specimens is able to diagnose measles with recent symptom onset while serology is an important tool for patients who are later in their clinical course or for purposes of retrospective case identification, as measles virus IgM typically develops up to one week after rash onset. Isolation of measles virus has typically been from respiratory, blood or urine specimens, but identification from the skin has not been documented. We describe the first known case of measles vaccine-associated disease in a patient on TNF inhibitor therapy in which genotype A Edmonston vaccine strain virus was identified from skin scrapings of the patient’s rash.

Case report

A 12 year-old boy was admitted to our hospital with a two-day history of tactile fevers, sore throat, rash, and conjunctivitis. His past medical history was remarkable for DiGeorge syndrome, repaired Tetralogy of Fallot, and juvenile idiopathic arthritis (JIA) diagnosed at 15 months of age, which had been controlled for the past 5 years with weekly injections of the tumor necrosis factor (TNF) inhibitor etanercept. Ten days earlier, he had inadvertently received the measles, mumps, rubella and varicella (MMRV) vaccine during a well child visit. The day prior to admission he developed a tactile fever, sore throat, and a pruritic rash that started on his neck as “spots and bled into one another.” Mild conjunctivitis developed the following day.

With the exception of live virus vaccines, he was otherwise up-to-date with his immunizations. He had neither traveled nor had any recent international visitors. He had recently undergone immune evaluation with a reported lymphocyte count of 1,100×10⁹ cells/L, of which the lymphocyte subsets were normal. His immune globulins were normal except for an elevated IgG₃ 140 mg/dL (normal 13 to 85 mg/dL), and a slightly elevated IgA at 242 mg/dL (normal 45 to 235 mg/dL). The lymphocyte mitogen and antigen studies were performed and included low-normal responses to Candida and phytohemagglutinin and normal responses to tetanus, concanavalin A, and pokeweed mitogen. Antibody titers to prior inactivated vaccines were normal.

After he developed the tactile fevers, his mother held further doses of etanercept, and he returned to his primary care provider. After a phone consultation with the Pediatric Infectious Diseases service, he was directly admitted to the hospital for evaluation and management of potential vaccine-associated disease. On admission he was afebrile (axillary temperature of 98.2 °C), with a heart rate of 91 beats/minute, blood pressure of 96/56 mmHg, respiratory rate of 20 breaths/minute, and oxygen saturation of 97%, and he appeared well. His examination was remarkable for mild conjunctivitis, a diffuse erythematous, blanching morbilliform rash that was present behind the ears, on the face, neck, back, chest, arms and legs, and palatal petechiae with erythema of the posterior pharynx. Mild post auricular cervical lymphadenopathy was present. There were no Koplik spots and no vesicular lesions. Due to the possibility of an atypical varicella rash in a patient on a biologic immunomodulatory medication, intravenous acyclovir 10 mg/kg IV every 8 hours was initiated.

Testing for group A Streptococcus, human herpes virus-6, and Epstein-Barr virus were sent upon admission and negative. His CBC showed a WBC count of 2.80 ×10⁹ cells/L with 53% neutrophils, 25% lymphocytes, 16.5% monocytes, and 5% eosinophils. His ALT was 34 U/L and AST 51 U/L. The Minnesota Department of Health was notified regarding concern for possible vaccine-associated disease from measles or varicella. Scrapings of the rash were sent to evaluate for varicella zoster virus, and a urine sample and a buccal swab were collected for evaluation of measles virus. Due to the recent exposure, measles and varicella IgM serologies were not sought.

As results were not yet available, he was discharged 24 hours after admission on oral acyclovir 800 mg every 6 hours for an additional 5 days with instructions to avoid school and other public situations until resolution of the rash. Two subsequent doses of etanercept were held while he was recovering from his illness. His rash and other symptoms resolved, although he had a brief hospitalization a week after discharge for hypocalcemic muscle contractions believed to be unrelated to his recent illness. No contacts of this case with similar immunodeficiency were identified and no similar cases were reported.

Materials and Methods

Nucleic acids were isolated from the urine and skin samples using the Qiagen Viral RNA Mini Kit. Detection of measles virus was attained with a real-time TaqMan RT-PCR targeting the nucleoprotein (N) gene [2]. Genotyping was determined following the WHO recommended target of sequencing the 450 nucleotides coding for the carboxy-terminal 150 amino acids of the N gene. Beckman Coulter’s CEQ 8800 and Dye Terminator Cycle Sequencing Quick Start Kit were used to obtain the sequence. Contiguous consensus and high quality base calls were guaranteed by inspection with Sequencher software. Reference sequences, designated by the CDC in affiliation with the World Health Organization’s LabNet program, were aligned with sample sequences using MEGA5 software. Phylogenetic analysis to establish genotype was performed by creation of a maximum parsimony tree [3].

Results

PCR of the skin sample was negative for varicella. The buccal swab was an inconclusive result based on extraction control failure. Both urine and skin samples were positive in triplicate for measles virus by RT-PCR and phylogenetically clustered as genotype A with Edmonston reference strain (AF266288) [Figure 1]. The urine and skin sample sequences had 100% matched identity to each other, as well as to the Moraten vaccine strain (U01999).

Figure 1.

Maximum parsimony analysis of taxa showing the phylogenetic relationships among our clinical isolates and reference sequences of other measles strains. Urine and skin samples were positive for measles virus by RT-PCR with 100% matched identity to each other and phylogenetically clustered as genotype A with Edmonston reference strain (AF266288). References sequences are from the WHO Measles Surveillance LabNet program/CDC.

Discussion

Measles can cause severe adverse events in immunocompromised patients, predominantly pneumonitis but also encephalitis. Much rarer is vaccine-associated measles, which has been reported in immunocompromised patients with HIV and severe combined immunodeficiency [4, 5].

The prototype for several of the live, attenuated measles vaccines now available, the Edmonston B strain was first isolated in 1954 from a child with measles and then adapted to growth conditions in nonpermissive cell cultures such as human kidney cells [6]. The original Edmonston B vaccine strain had an unacceptably high rate of fever and rash, but further attenuation led to the Edmonston-Moraten strain, which became the seed strain available in the United States [7].

Live virus vaccines are not necessarily contraindicated in patients with immunodeficiencies such as DiGeorge syndrome, nor in those with autoimmune diseases such as JIA; and some such patients have tolerated measles vaccine without adverse effects [8]. In this instance MMRV vaccine was contraindicated given the recent use of etanercept, which is a soluble TNF receptor:Fc fusion protein. Etanercept inhibits components of cellular and humoral immunity allowing for the attenuated measles vaccine strain to cause more significant symptoms associated with wild-type measles infection. Due to their immunomodulatory effects, administration of live vaccines is contraindicated in patients on these drugs.

As measles incidence has decreased in developed countries after introduction of measles vaccination, PCR-based testing of throat and urine specimens has become the primary method of diagnosis in many clinical settings. In our patient’s case, excretion of measles virus in his urine enabled detection and sequencing of the virus as a measles vaccine strain. However, as vaccinated, asymptomatic children can shed measles virus RNA in the urine for up to two weeks after vaccination [9], the diagnosis of vaccine-associated disease was determined from the skin scrapings. Though it is known that measles virus is present in the skin rash [10], to our knowledge this is the first reported case of measles in which the measles vaccine strain was detected and genotyped from the skin rash. While this could be due to the underlying immune system abnormality associated with DiGeorge syndrome, this patient’s immune system was not significantly impaired based on recent testing. The ability to detect measles virus from the skin raises the possibility of adding this test to future investigations of vaccine-associated disease in patients with a rash.