Seroprotection after measles vaccination in children with sickle cell anemia receiving hydroxyurea Authors: Alexandra Power-Hays, Abigail Kincaid, Kathyrn E. McElhinney, Teresa S. Latham, Charles T. Quinn, Robert W. Frenck, Russell E. Ware Published: 2026 ABSTRACT Key Points•Children with SCA have similar post–measles vaccine seroprotection compared with the general population.•Hydroxyurea at the time of vaccination does not diminish measles seroprotection, and earlier hydroxyurea initiation enhances protection. Introduction Measles is a highly contagious, yet vaccine-preventable, viral infection with increasing transmission across the United States and a persistent endemic disease globally, particularly in sub-Saharan Africa and India.1, 2, 3 Infections are often mild with cough, coryza, and conjunctivitis, but can result in permanent neurologic disability, blindness, hearing loss, and death. Most healthy children achieve seroprotection after measles vaccination; however, several factors diminish seroconversion rates, including age at immunization, nutritional status, and concurrent infections.4,5 People with sickle cell anemia (SCA) develop functional asplenia early in life, and many develop immune dysregulation and altered humoral immunity.6,7 Splenic dysfunction increases the risk of bacteremia with encapsulated organisms, but individuals with SCA also exhibit diminished antibody responses after certain vaccines, including those against pneumococcus, influenza, and hepatitis B.7, 8, 9, 10 It is currently unknown whether children with SCA develop adequate protection after measles vaccination. Increased vaccine refusal across the United States may cause a decline in herd immunity, so individual seroprotection becomes of even more important.1, 2, 3,11 Furthermore, there is an US Food and Drug Administration warning that hydroxyurea, the primary disease-modifying treatment for SCA, impairs antibody responses to measles vaccination based on a secondary analysis of only 14 hydroxyurea-treated infants in the phase 3 BABY HUG trial (ClinicalTrials.gov identifier: NCT00006400).12 General clinicians, public health officials, and families may worry about the impact of coadministration of measles vaccines and hydroxyurea, particularly in areas in which measles is endemic.13 With ongoing outbreaks and the global colocation of high SCA burden and measles, it is critical to determine whether this population is adequately protected using immunization schedules designed for the general population, and whether hydroxyurea affects the measles antibody response. Methods We conducted a retrospective cohort study to determine the rates of seroprotection after routine measles vaccination in children with SCA enrolled in TREAT (ClinicalTrials.gov identifier: NCT02286154), a prospective trial evaluating pharmacokinetic-guided hydroxyurea dosing in people aged 0.5 to 21.0 years with hemoglobin SS or hemoglobin S/β0 thalassemia in Cincinnati, OH. Samples up to 18 years after vaccination from participants whose families signed consent for future research were included. Participants without documented measles vaccination were excluded. This study was approved by the Cincinnati Children’s Hospital institutional review board and was conducted in accordance with the Declaration of Helsinki. Serum or plasma samples stored at −80°C were tested for anti–measles immunoglobulin G levels using a commercially available enzyme-linked immunoassay (El 2610-9601G; Euroimmun, Lübeck, Germany). Specimens were selected via convenience sampling based on sample availability rather than at specified intervals after vaccination. The primary study outcome was the proportion of participants achieving antimeasles seroprotection after first measles vaccination (MV1) or second measles vaccination (MV2). Seroprotection was defined as a corrected optical density (OD) of ≥1.1, corresponding to an immunoglobulin G titer of ≥200 IU/L and to a plaque neutralization titer of 120 mIU/mL.14 Secondary outcomes included the effects on seroprotection of age, hydroxyurea treatment at vaccination, and reduced splenic filtrative function, defined as an erythrocyte pit count of ≥5.0% within 1 year of vaccination. Proportions, means, and medians were compared via the Fisher exact test, Student t test, and Mann Whitney U test, respectively. The impact of clinical and demographic variables on seroprotection was determined via logistic regression and on OD via linear regression. Variables with P values of ≤.2 in univariate analyses were included in multivariable regression models. Statistical significance was defined as P value <.05. Results A total of 68 TREAT participants had stored serum or plasma samples, of whom 66 (97%) had documented measles vaccination. Among 35 participants evaluated after MV1, 33 (94%) had seroprotective titers, and among 50 participants evaluated after MV2, 46 (92%) had seroprotective titers (Table 1).Table 1.Clinical characteristics of children with SCA based on seroprotection status after MV1 and MV2SeroprotectionNo seroprotectionP valueMV1, n (%)n = 33 (94.3)n = 2 (5.7) Time after vaccination, median (IQR), y0.8 (0.5-1.9)0.5 (0.3-0.7).3 Sex, n (%)1.0 Female13 (93)1 (7) Male20 (95)1 (5) Age at vaccination, median (IQR), y1.1 (1.0-1.4)5.3 and 19.3— Hydroxyurea at time of vaccination, n (%).5 Yes17 (100)0 (0) No16 (89)2 (11) Pit count available, n (%)29 (88)2 (100)1.0 Elevated pit count, n/available (%)9/29 (31)1/2 (50)1.0MV2, n (%)n = 46 (92)n = 4 (8) Time after vaccination, median (IQR), y1.7 (0.5-4.2)7.2 (4.0-9.8).1 Sex, n (%)1.0 Female22 (92)2 (8) Male24 (92)2 (8) Age at vaccination, mean (SD), y4.5 (0.8)5.6 (1.1).01 Hydroxyurea at time of vaccination, n (%).6 Yes33 (94)2 (6) No13 (87)2 (13) Age at hydroxyurea initiation, mean (SD), y3.5 (3.9)8.9 (5.2).01 Pit count available, n (%)42 (91)4 (9)1.0 Elevated pit count, n/available (%)25/42 (60)4/4 (100).3 Pit count, mean (SD), %12.9 (12.4)23.6 (12.3).1Seroprotection refers to antibody titers of ≥200 IU/mL.Bold text indicates statistically significant p-value ≤ 0.05.IQR, interquartile range; SD, standard deviation. Table 1.Clinical characteristics of children with SCA based on seroprotection status after MV1 and MV2SeroprotectionNo seroprotectionP valueMV1, n (%)n = 33 (94.3)n = 2 (5.7) Time after vaccination, median (IQR), y0.8 (0.5-1.9)0.5 (0.3-0.7).3 Sex, n (%)1.0 Female13 (93)1 (7) Male20 (95)1 (5) Age at vaccination, median (IQR), y1.1 (1.0-1.4)5.3 and 19.3— Hydroxyurea at time of vaccination, n (%).5 Yes17 (100)0 (0) No16 (89)2 (11) Pit count available, n (%)29 (88)2 (100)1.0 Elevated pit count, n/available (%)9/29 (31)1/2 (50)1.0MV2, n (%)n = 46 (92)n = 4 (8) Time after vaccination, median (IQR), y1.7 (0.5-4.2)7.2 (4.0-9.8).1 Sex, n (%)1.0 Female22 (92)2 (8) Male24 (92)2 (8) Age at vaccination, mean (SD), y4.5 (0.8)5.6 (1.1).01 Hydroxyurea at time of vaccination, n (%).6 Yes33 (94)2 (6) No13 (87)2 (13) Age at hydroxyurea initiation, mean (SD), y3.5 (3.9)8.9 (5.2).01 Pit count available, n (%)42 (91)4 (9)1.0 Elevated pit count, n/available (%)25/42 (60)4/4 (100).3 Pit count, mean (SD), %12.9 (12.4)23.6 (12.3).1Seroprotection refers to antibody titers of ≥200 IU/mL.Bold text indicates statistically significant p-value ≤ 0.05.IQR, interquartile range; SD, standard deviation. Clinical characteristics of children with SCA based on seroprotection status after MV1 and MV2 Seroprotection refers to antibody titers of ≥200 IU/mL. Bold text indicates statistically significant p-value ≤ 0.05. IQR, interquartile range; SD, standard deviation. Impact of age at vaccination Older age at measles immunization was associated with reduced seroprotection and antibody titers. In the MV1 cohort, the 2 children who did not achieve seroprotection were older at MV1 (aged 5.3 and 19.5 years) than those who were seroprotected (median age, 1.1 years [interquartile range, 1.0-1.4]; Table 1). In the MV2 cohort, the mean (±standard deviation) age at vaccination of those who were seroprotected was 4.5 (±0.8) years compared with 5.6 (±1.1) years in those not seroprotected (P = .01). Via logistic regression, each additional year at MV2 was associated with decreased odds of seroprotection (odds ratio, 0.3; P = .02; Table 2). Similarly, increased age at vaccination was associated with lower antibody titers reaching statistical significance after MV1 (P = .03; Table 3) and trending toward significance after MV2 (P = .09; Table 2). Notably, of the 4 children who were not seroprotected after receiving 2 measles vaccines, 2 were also delayed in receiving MV1.Table 2.Logistic regression analyses to evaluate the impact of clinical factors on seroprotection after MV2Univariate logistic odds ratio (95% CI), P valueMultivariable logistic odds ratio (95% CI), P valueAge at MV2 (y)0.3 (0.08-0.8), P = .020.3 (0.08-1.2), P = .1On hydroxyurea at MV2 (yes/no)0.4 (0.05-3.1), P = .4—Pit count (continuous)0.9 (0.9-1.0), P = .10.9 (0.9-1.1), P = .6CI, confidence interval.Table 3.Linear regression analyses to evaluate the impact of clinical factors on measles antibody titers after MV1 and MV2Univariate linear beta (95% CI), P valueMultivariable linear beta (95% CI), P valueAge at MV1, y−0.1 (−0.2 to −0.03), P = .01−0.1 (−0.2 to −0.01), P = .03On hydroxyurea at MV1 (yes/no)0.07 (−0.6 to 0.8), P = .8—Pit count, %−0.03 (−0.09 to 0.02), P = .2−0.03 (−0.08 to 0.02), P = .2Age at MV2, y−0.2 (−0.5 to 0.04), P = .09−0.2 (−0.5 to 0.03), P = .09On hydroxyurea at MV2 (yes/no)0.2 (0.04-0.80), P = .030.2 (−0.02 to 0.8), P = .06Pit count (continuous)−0.01 (−0.03 to 0.02), P = .6—CI, confidence interval. Table 2.Logistic regression analyses to evaluate the impact of clinical factors on seroprotection after MV2Univariate logistic odds ratio (95% CI), P valueMultivariable logistic odds ratio (95% CI), P valueAge at MV2 (y)0.3 (0.08-0.8), P = .020.3 (0.08-1.2), P = .1On hydroxyurea at MV2 (yes/no)0.4 (0.05-3.1), P = .4—Pit count (continuous)0.9 (0.9-1.0), P = .10.9 (0.9-1.1), P = .6CI, confidence interval. Logistic regression analyses to evaluate the impact of clinical factors on seroprotection after MV2 CI, confidence interval. Table 3.Linear regression analyses to evaluate the impact of clinical factors on measles antibody titers after MV1 and MV2Univariate linear beta (95% CI), P valueMultivariable linear beta (95% CI), P valueAge at MV1, y−0.1 (−0.2 to −0.03), P = .01−0.1 (−0.2 to −0.01), P = .03On hydroxyurea at MV1 (yes/no)0.07 (−0.6 to 0.8), P = .8—Pit count, %−0.03 (−0.09 to 0.02), P = .2−0.03 (−0.08 to 0.02), P = .2Age at MV2, y−0.2 (−0.5 to 0.04), P = .09−0.2 (−0.5 to 0.03), P = .09On hydroxyurea at MV2 (yes/no)0.2 (0.04-0.80), P = .030.2 (−0.02 to 0.8), P = .06Pit count (continuous)−0.01 (−0.03 to 0.02), P = .6—CI, confidence interval. Linear regression analyses to evaluate the impact of clinical factors on measles antibody titers after MV1 and MV2 CI, confidence interval. Impact of hydroxyurea treatment Hydroxyurea treatment at the time of vaccination was associated with equivalent or improved postvaccine seroprotection (Table 1). Among 17 children (49%) taking hydroxyurea at MV1, all 17 (100%) achieved seroprotection compared with 16 of 18 (89%) who were not receiving hydroxyurea (P = .5). Among 33 children (69%) taking hydroxyurea at MV2, 31 (94%) achieved seroprotection compared with 13 of 15 (87%) not taking hydroxyurea (P = .4). Furthermore, among those on hydroxyurea at MV2, earlier hydroxyurea initiation was associated with improved seroprotection, with each year younger at treatment initiation associated with more than a fourfold increase in the odds of seroprotection (odds ratio, 4.4; 95% confidence interval, 1.2-14.3; P = .02; Table 2). Factors not associated with seroprotection Participant sex did not affect seroprotection (Table 1) or antibody titer. Similarly, the erythrocyte pit count did not affect seroprotection (Tables 2 and 3). Longitudinal assessment of repeat vaccination and antibody persistence A total of 19 participants had samples available both after MV1 and MV2. Of these, 18 (94.7%) reached seroprotection after MV1, and 17 (89.5%) remained seroprotected after MV2. One child was below seroprotective thresholds after both MV1 and MV2; this child was immunized in accordance with national recommendations at 1.8 and 4.4 years of age, initiated hydroxyurea treatment at 2.7 years of age, and had an elevated pit count of 22.6% at treatment initiation. The measles antibody titer from another child was initially robust, but declined below the seroprotective threshold after MV2. This child was on a catch-up schedule due to recent immigration, receiving MV1 at the age of 4.2 years, followed by MV2 at the age of 4.7 years. The child initiated hydroxyurea treatment at the age of 4.3 years, and had an elevated pit count of 29.5% at treatment initiation. Serial post-MV2 samples were available for 29 participants (median interval, 2.9 years [range, 0.3-6.7]), of whom 27 (93%) had a decrease in the antibody titer over time. Among these children, the mean (±standard deviation) rate of OD decline was 3.7% (±3.7%) per year, similar to the general population. Despite this expected modest decline, most participants maintained measles antibody titers above the seroprotective threshold. Discussion Overall, a high proportion of children with SCA achieved seroprotection in response to measles immunization. The observed seroprotection rates of 94% after MV1 and 92% after MV2 are comparable to the 93% to 96% and 97% to 100%, respectively, achieved in the general population.15, 16, 17 The slightly lower seroprotection rates after MV2 than MV1 differ from the pattern of increased response to MV2 in the general population and warrant further evaluation; however, these proportions far exceed the levels of measles seroprotection achieved in profoundly immunocompromised populations, such as children with HIV (57% seroprotection)18 or those on immunosuppression after solid organ transplantation (50% seroprotection).19 Older age at the time of measles vaccination was a risk factor for decreased seroprotection and diminished antibody titers in this cohort. Although our small sample size precludes definitive conclusions, the mechanisms by which SCA may impair immunogenicity are multifactorial and include loss of splenic marginal zone B cells, impaired switching of memory B cells, reduced helper T cells, or chronic immune activation leading to T-cell exhaustion.7,20 In our study, an increased erythrocyte pit count, which reflects reduced splenic filtrative function, was not associated with seroprotection or antibody titers. This observation distinguishes 2 key functions of the spleen: blood filtration in the red pulp and immunological responses in the white pulp. Loss of splenic filtrative function is one of the earliest manifestations of SCA damage6,7,21 but likely occurs earlier than impaired adaptive immunity and atrophy of the spleen parenchyma. Based on these data, families should be counseled about the potential for diminished measles protection with a delayed vaccine schedule, although further investigation in larger cohorts is necessary. Younger age at hydroxyurea initiation was associated with fourfold increased odds of seroprotection per year in this study. The US Food and Drug Administration formally warns that hydroxyurea decreases a patient’s antibody response to measles vaccination, but this was based on a small secondary analysis of the BABY HUG trial that found diminished early antibody responses after MV1 in 14 children on hydroxyurea compared with 24 on placebo.12,22 Notably, in the aforementioned trial, many of the children with low measles antibody titers were tested <2 weeks after immunization; hence, the peak response to vaccination was likely underestimated. In our larger analysis, hydroxyurea did not reduce antibody titers or seroprotection after MV1 or MV2. Critically, our results demonstrate that patients should be prescribed hydroxyurea without concern for diminished antibody response and do not support avoiding or delaying hydroxyurea treatment due to measles vaccination in children with SCA. Although older age at vaccination and at hydroxyurea initiation were associated with diminished initial seroprotection in a small number of children with SCA, once an adequate immune response was established, waning of measles antibody titers occurred at a rate of 3.7% per year, comparable to 4.8% reported in healthy populations.17,23 Most individuals who do mount an initial antimeasles response, the durability of that immunity is likely due to the T-cell dependent generation of long-lived plasma cells.24 This contrasts with the pneumococcal polysaccharide vaccine, which elicits a T-cell–independent response that relies more heavily on splenic marginal zone B cells and is associated with accelerated antibody decay and the need for booster vaccines to maintain seroprotection in children with SCA.20,25, 26, 27 Our findings do not suggest that children with SCA require a measles booster any more urgently than the general population.28 Limitations of this study include its retrospective design, use of convenience samples, and small sample size. Because samples were not timed around vaccination, there was a trend toward longer intervals between MV2 and sample collection among those without seroprotective titers, suggesting some children may have had adequate titers at earlier time points. A prospective, multinational study that includes children in the United States, Ghana, and Tanzania (Verifying Antibodies after Live Immunization Delivery, ClinicalTrials.gov identifier: NCT07356050) with larger sample size and timed antibody testing is now underway to address these limitations. An additional limitation is that seroprotection refers to an antibody titer threshold associated with reduced infection, but it does not necessarily guarantee infection prevention.14 Conversely, many people who have been immunized will mount an increased antibody response upon exposure to measles virus24; real-world studies are necessary to evaluate measles infection risk in people with SCA. In conclusion, children with SCA have similar seroprotection rates after measles vaccination as those of the general population, but increased age and delayed hydroxyurea initiation were both associated with decreased antibody responses. Loss of splenic function may contribute, but this analysis did not find a statistically significant association with splenic filtration. Critically, hydroxyurea treatment was not associated with diminished antimeasles seroprotection rates or antibody titers. Rather, earlier hydroxyurea initiation was protective and associated with improved vaccine responses. Measles vaccination and hydroxyurea treatment are both critical interventions to prevent morbidity in this vulnerable population; we recommend that all infants and young children with SCA receive hydroxyurea and also receive measles vaccinations on time to protect against this highly contagious viral illness. Conflict-of-interest disclosure: A.P.-H. and R.E.W. were formerly on a health equity advisory board for Novo Nordisk. R.E.W. receives research donations from Theravia and Hemex Health; serves as a medical adviser to Nova Laboratories, Theravia, and Hemex Health; and serves on a data and safety monitoring board for a clinical trial sponsored by Vaxart. The remaining authors declare no competing financial interests.