APMHC

Introduction

Human Cytomegalovirus (CMV) is the most common cause of congenital infections. There are two primary reasons for this: 1) CMV is the only pathogen capable of infecting the fetus during both primary and non-primary (recurrent) maternal infections; 2) the virus can be shed for years from children with infection acquired prenatally or in the first years of life. Congenital CMV infection occurs in approximately 0.5%-2% of all liveborn neonates, but in about 40% of infants born to mothers with primary infection [1]. In these women, the viral transmission rates are increasingly associated with the progression of pregnancy and placental aging, while the symptomatology of the infected infants is inversely correlated with the weeks of pregnancy [2]. When transmitted in the first three months of pregnancy, because of the high susceptibility of the embryonal cells to CMV and the lack of maternal and fetal immune defenses against the virus, CMV can cause severe neurological and sensorineural sequelae up to 40% in infected newborns [3,4].

Primary CMV infection is asymptomatic in most of pregnant women. A flu-like syndrome with lymphocytosis and elevated amninotransferases occurs in some patients [5]. Due to the lack of a universal screening for CMV in pregnancy, these manifestations are underestimated and the diagnosis may be missed. In many countries, however, CMV screening has gained increased importance (notably, universal screening was introduced in Italy in December 2023) and therapeutic approaches have been reported for years [6-14]. Hyperimmune globulin (HIG) administration to pregnant women with primary CMV infection was initiated over 25 years ago. Several controlled or observational studies reported an excellent safety and efficacy profile for the prevention of congenital CMV disease using high dosage (HD: 200 units/kg) of HIG [6-8,11,12]. Less clear is the efficacy profile to prevent motherto-fetus CMV transmission using monthly HIG administration of 100 units/kg [7,9,14].

Since the antiviral and immunomodulatory efficacy of HIG appeared to be higher if given soon after maternal infection, and there was no significant difference between the number and the frequency of HIG infusions, we performed a long-term cohort study by giving only one timely HD-HIG infusion to 76 women with primary CMV infection in the first trimester. This dose was used because: 1) The majority of the patients enrolled were treated 6 weeks after their presumed maternal viremia and fetal CMV infection could have already occurred, thus HD-HIG infusion might have been therapeutic rather than preventive; 2) A single HD-HIG infusion has been previously reported to be therapeutic in pregnant women with fetal CMV disease [7]. First aim was to prevent the development of congenital CMV disease, mostly including neurological and hearing disabilities. Other objectives were prevention of mother-to-fetus CMV transmission, safety, and cost-benefit ratio. As controls 80 women, who did not accept to be treated with HIG, were followed. All the infants, from mothers HD-HIG treated or not, were followed up for a mean of 4.6 years.

Materials and Methods

Study Design

From July 1, 2017 to June 30, 2022, 1246 pregnant women with suspected CMV infection asked for consultation the first author by mails, messages and calls, including 168 (13.5%) with confirmed primary CMV infection in the first trimester. CMV infection was first determined by detection of CMV IgG and IgM in diagnostic laboratories then confirmed by detection of antibodies and IgGavidity, before HD-HIG infusion or alternative therapy. Detection of CMVDNA in blood (DNAemia) was also suggested but not performed by all women. Twelve (7.1%) women decided to interrupt their pregnancy before 12 weeks of gestation. All 156 (92.9%) women, who continued their pregnancy, were suggested to receive one infusion of HD-HIG for prevention of fetal CMV infection or disease in Villa Mafalda Clinic, Rome, Italy, 4 to 9 weeks after the presumed maternal infection (MI), for the prevention of fetal CMV infection or disease. None patient had HIV infection or immunosuppressive disease or any other current disease. The treatment group was self-selected; the decision to receive HD-HIG was made by each patient following counselling with her physician.

For each patient essential data were: maternal age at conception, gestational age at the time of maternal CMV infection, gestational age at the HD-HIG infusion or date of consultancy or start of alternative therapy, gestational age at the time of CMV DNA detection in blood, viral load in the amniotic fluid, prenatal manifestations of CMV disease, gestational age at delivery, neonatal birth weight, clinical and laboratory abnormalities in the infants with congenital infection, and follow-up for at least two years.

The gestational age at maternal infection was estimated after seroconversion as half way between the last seronegative and the first seropositive serum or the beginning of symptoms (i.e. fever, flu-like syndrome) or laboratory abnormalities (i,e. elevated transaminases, lymphocytosis), when these were reported [5]. For women who had both IgM and IgG antibodies associated with a very low avidity in their first serological testing, being CMV negative in a previous pregnancy and having a child under 3 years attending a nursery, the date of maternal infection was estimated to be 4 weeks prior to testing to account for the window of viremia and subsequent IgM/IgG production from B-lymphocytes. If only CMV IgM antibodies were detected, and followed by IgG detection, the maternal infection was anticipated of 2 weeks. Maternal infection was considered to have occurred just after conception. In mothers with positive IgG and IgM within 6 weeks of pregnancy, maternal infection was considered to occur soon after conception, because of the natural immune depression aimed to prevent fetal rejection since the fetus is like a 50% allograft [3].

Amniocentesis was suggested at 19-20 weeks gestation for detection of CMV DNA in the amniotic fluid. If amniotic fluid was positive, regardless of fetal abnormalities by ultrasound (US), fetal Magnetic Resonance Image (MRI) was suggested to at 20-21 weeks to exclude cerebral malformations and offer the option of legal termination of pregnancy (TOP), which is within 22 weeks in Italy. All neonates were tested for CMV IgG, IgM, and DNA in blood and/or urine. Fetal CMV infection was considered symptomatic if the fetal US scan showed ventriculomegaly or echodensities in the brain, bowel or liver, or if the brain MRI revealed abnormalities like leukodystrophy, cerebral and cerebellar dystrophy or neuronal migrational disorders [15]. Intrauterine growth restriction was defined as head and abdominal circumferences that were below the 10th percentile. All infected infants received brain US and MRI scan, abdominal US, auditory brain-stem evoked response (ABR) studies, and an ophthalmoscopy evaluation. CMV-infected infants were monitored for 2 to 8 years by routine clinical evaluation, sensorineural hearing and eye examinations.

All HD-HIG treated patients gave written informed consent for receipt of hyperimmune globulin infusions. All patients provided consent permitting the anonymous use of their medical records. The study was approved by the Internal Review Board of the University of L’Aquila, Italy (protocol number: 37/109, March 14, 2009), which covered the years of patient enrolment and followup, and by the Ethical Committee of the non-profit Association Mother-Infant Cytomegalovirus Infection (AMICI; Registration No. 220615).

Laboratory Diagnosis of CMV Infection

CMV-specific antibodies were detected by enzyme immunoassays from Diesse Diagnostica Senese (Siena, Italy) and DiaSorin (Saluggia, Italy). CMV IgG avidity was examined using the diagnostic kit (low <0.15%, high >0.25%) from Bouty (Milan, Italy). CMV DNA genome copies/ml were detected by RealTime PCR from Amplimedical - Bioline (Turin, Italy) and Qiagen (Hoffmann LaRoche). Both serological and virological testing was performed according to manufacturer’s instructions.

Hyperimmune Globulin

Commercial hyperimmune globulin (Cytomegatect Biotest, Germany) was given at the dosage of 200 Paul Ehrlich units/kg of maternal weight. To avoid immediate or late side effects related to the immunotherapy, the infusion speed was gradually increased from 20 ml/h in the first 15 minutes, to 30 ml/hour in the minutes 16-30, to 40 ml/h in the minutes 31-45, and to 50 ml/h until the end of the infusion. The duration of the infusion was related to the HD-HID dosage, then to the weight of the pregnant women, who were dismissed after 30 minutes.

CMV Disease

Symptomatic congenital CMV disease was defined as neurological involvement, including microcephaly (head circumference below the 5th percentile for gestational age), periventricular calcifications, cerebral or cerebellar atrophy, cerebral abnormalities (polymicrogyria, lissencephaly, pachygyria), leukodystrophy, ventricular and subependymal abnormalities, seizures in an infant with CMV DNA in cerebrospinal fluid, full or partial hearing loss in one or both ears, chorioretinitis, and purpura with or without thrombocytopenia. The infants were considered asymptomatic when mild hepatosplenomegaly or US findings resolved over the first weeks of life.

CMV disease at ≥ two year of age was defined by mental retardation (speech delay, repetitive behaviors, social withdrawal, intense outburst, IQ below 70 in children aged ≥6 years) or moderate to severe motor delay, and auditory or visual impairment. The threshold by ABR for normal hearing was defined as 0 to 20 dB; abnormal responses were defined as mild (threshold, 21 to 45 dB), moderate (threshold, 46 to 70 dB), or severe (threshold, at least 71 dB).

Statistical Analysis

Qualitative variables were expressed as the number of cases or percentages. Differences between groups for qualitative variables were assessed using Pearson’s chi-squared test or Fisher’s exact test for small frequencies. For the multivariable logistic regression, the model included maternal age, gestational age at infection, and timing of HD-HIG administration as pre-specified covariates to control for potential confounding. Continuous variables were represented as mean and standard deviation (SD). Parameters were tested for normality with Shapiro-Wilk test and the rank-sum test or t-test were used to compare data between groups as appropriate. Multivariate logistic regression analysis was conducted to identify independent variables correlated with CMV infection or disease in fetuses and newborns. For each analysis, an alpha level of 0.05 was considered to be statistically significant. The statistical analysis was performed using the STATA/BE 18 software for Windows.

Results

Enrolment and Data: As reported in Table 1, 76 pregnant women (two had a twin pregnancy) accepted to be treated with only one HD-HIG infusion at a mean gestational week of 14.4 (range: 8-21), after 6.8 mean gestation week from the presumed maternal infection. Seroconversion occurred in 43 patients (56.6%). All but five pregnant women were Italian and were assisted during the infusions by G.N. None had any HD-HIG-associated side effects during the infusion or later: all were in contact with the author by cell phone. As controls 81 women (one had a twin pregnancy), who did not accept to be treated with HD-HIG, were followed-up from a mean gestational week of 15.1 (range: 7-22), after 7.5 mean gestation weeks from the maternal infection. Of these women, 51 (63.75%) decided to have no therapy, 20 (25%) received non-specific immunoglobulin, 9 (11.25%) were treated with valacyclovir.

                         No. of subjects who delivered by Caesarean         16                  23 (30.7)

                                section (%)                                                        (22.9)

MRI: Magnetic Resonance Image US: ultrasound; VAC: valacyclovir

Table 1: Univariate and Multivariate (Logistic Regression) analysis of possible predictors of CMV infection or disease in fetuses and newborns.

Mother-to-fetus CMV transmission and disease: As shown in Table 1, 13 of 56 HD-HIG-treated mothers (23.2%) had positive CMV DNA in the amniotic fluid. Six women (46.1%), one of whom had ultrasound fetal abnormalities, requested TOP. Of 72 neonates, 9 (12.5%) were CMV positive. Totally, CMV was transmitted to 15/78 fetuses/neonates (19.2%). Notably, in the subgroup of 20 patients treated within 5 weeks of the presumed maternal infection, the vertical transmission rate was 0%. This reduction was statistically significant when compared to patients treated after 5 weeks (0% vs. 25.9%; p<0.013). Clinical/image abnormalities occurred in 2 neonates (17%), both with left deafness which persisted after valganciclovir therapy. The longterm outcome showed that all children had a normal psychomotor development (Tables 1 and 4).

Among the control women, 11 of 57 (19.3%) had positive CMV DNA in the amniotic fluid: 4 women (30.8%) opted for TOP, and 2 of them showed fetal ultrasound/MRI abnormalities. Two intrauterine fetal deaths, and one stillborn also occurred (Table1). Of 74 alive neonates, 20 (27%) were CMV-positive and 11 of them (55%) were symptomatic. Totally, CMV transmission occurred in 27/81 fetuses/neonates (33.3%), being significantly higher (p=0.044) than those from HD-HIG-treated women. In particular, maternal-fetal CMV transmission was significantly higher (40% vs 0%; p=0.002) in the control women than in HD-HIG-treated women within 5 weeks from the maternal infection (Table 2). Moreover, it is statistically significant the high number of fetuses/ neonates with CMV diseases from control women compared to HD-HIGtreated mothers (59.2% vs 20%; p=0.003). Of 11 CMVinfected symptomatic neonates, 10 (90.9%) were treated with valganciclovir: 9 (81.8%) had neurological abnormalities and/ or abnormal hearing (Table 5). In particular, the difference in the occurrence of abnormal neurological outcome is highly significant (7 vs. 0; p=0.008).

TOP: Termination of pregnancy; IUD: Intrauterine death

Table 2: Interval in weeks gestation (WG) between Maternal Infection and infusion of hyperimmune globulin (HIG) or controls (date of consultancy or beginning of therapy) as a possible predictor of congenital CMV infection or disease at birth and follow-up.

CMV DNAemia: As reported in Table 3, this testing was performed in 67 HD-HIG-treated women (88.1%), 41 of whom (61.2%) were positive and 13 (31.7%) transmitted CMV to their fetuses. Mean genome/ml values were 2.454 (range 24-14.047). Mother-tofetus CMV transmission was significantly lower (p=0.006) in the women who had negative CMV DNAemia at their first testing, since it occurred in only 1 of 26 (3.8%) women having negative CMV DNAemia but in 12 with positive DNAemia (32.5%).

Among controls, CMV DNAemia was detected in 57 women (71.25%), 35 of whom (61.4%) were positive. Mean genome/ml values were 848.8 (range 45-13.110). There was no statistically significant difference between non-HIG-treated patients having positive (36.4%) or negative (27.3%) DNAemia in mother-to-fetus CMV transmission (p=0.486). Conversely, there was a statistically significant difference (p=0.006) between HD-HIG-treated mothers having negative DNAemia (3.8%) and non-HIG-treated women having negative CMV DNAemia (27.3%).

IgG-avidity: As reported in Table 3, this test was performed before and after HD-HIG infusions in 53 subjects, and twice in 52 controls. There was an increase in the mean avidity values in both groups. Nevertheless, the increased avidity after HD-HIG infusion (from 0.10% to 0.24) was statistically significant (p=0.021), contrary to the raised avidity (from 0.11% to 0.17%) in the controls.

M/F: Maternal/Fetal; g/ml: genome copies/ml; *Statistical comparison between HIG-treated patients having positive or negative DNAemia (32.5% vs. 3.8%): p=0.006; ** Statistical comparison between non-HIG-treated patients having positive or negative DNAemia (36.4% vs. 27.3%): p=0.486

Table 3: Univariate and Multivariate (Logistic Regression) analysis of DNAemia and IgG avidity as possible predictors of congenital CMV infection/disease in fetuses and neonates.

Safety and Cost/Efficacy Ratio

HD-HIG infusions were safe in all patients, in spite of the fact that many of them came back to their cities from Rome in the same day of the infusion. The cost/efficacy ratio was fully achieved, since the mean cost of one HD-HIG infusion is 4.250 euros, which is lower than the cost of at least two infusions of non-specific immunoglobulin, valacyclovir therapy, and mostly of disabilities caused by congenital CMV infection.

Discussion

This prospective cohort study, featuring a robust long-term followup, showed that a single, timely HD-HIG infusion significantly prevents congenital CMV disease following first-trimester primary infection. During the study free non-specific immunoglobulin or valaciclovir were available in Italy for use in pregnancy, so a randomized trial with placebo was impossible. In spite of this

limitation, we observed significantly lower numbers of fetuses/ infants with CMV disease at birth from HD-HIG treated mothers than from control mothers (3.9% vs. 19.6%; p= 0.003). Moreover, although the non-randomized nature of the study may introduce selection bias, the stark contrast in transmission rates (0% in the early treatment group) and the long-term neurological outcomes provide compelling evidence for the efficacy of timely HD-HIG intervention.

Further, after long-term evaluation none of 72 children born to HD-HIG treated mothers developed neurological abnormalities, and only two had unilateral deafness. Also, the CMV transmission rate was significantly (p=0.044) decreased among all fetuses or infants born of HD-HIG-treated mothers compared to infants from controls. The transmission rate of CMV in non-HD-HIGtreated women was 33.3%, [95% CI: 26.2-45.0%]), which was significantly higher than in the control group (19.5%).

CMV: cytomegalovirus; F: female; g/ml: genome copies/ml; HIG: hyperimmune globulin; M: male; MRI: Magnetic Resonance Imaging; NP: not performed; SC: seroconversion; TOP: termination of pregnancy; US: ultrasound; VGC: valganciclovir; WG: weeks of gestation

Table 4: The outcome of pregnant women with primary CMV infection before 14 weeks’ gestation who were treated with hyperimmune globulin before amniocentesis to prevent maternal-fetal transmission.

Table 5: The outcome of pregnant women with primary CMV infection before 14 weeks’ pregnancy who were not treated with hyperimmune globulin (HIG) to prevent maternal- fetal transmission.