Letter to the editor: "hook" (prozone) effect in sars-cov-2 anti-spike binding antibody levels following vaccination, infection, or monoclonal antibody in solid organ transplant recipients.

Anti-spike antibody (Ab) assays were authorized to determine prior exposure to SARS-CoV-2 and have since been adopted by some clinicians to measure humoral response to vaccination and estimate vulnerability to infection.1, 2 As such, many commercial assays currently in use for clinical and research purposes have not undergone validation in the setting of high antibody concentrations which may be achieved through repeat antigen exposure or monoclonal antibody (mAb) therapies.3 This is noteworthy because immunoassays used in other settings have been observed to return imprecise or artifactually low quantitative results due to a “hook” or prozone effect in undiluted samples at their upper limits of quantification, that is, in the setting of very high binding antibody titers.4 We sought to determine whether a hook effect occurs in a semi-quantitative binding assay (Roche-Elecsys anti-SARS-CoV-2 S) which is widely used by medical providers and researchers, given potential clinical implications, particularly for vulnerable populations such as solid organ transplant recipients (SOTRs). Within a national observational cohort (JH IRB00248540) and a single-center clinical trial (NCT04969263) of vaccinated SOTRs, participants underwent Ab testing before and after receipt of either the mAb combination Tixagevimab/Cilgavimab (T/C) (observational cohort) or a third or fourth SARS-CoV-2 mRNA vaccine (clinical trial participants). Participants with either a paradoxical decrease or weak (<10-fold) rise in Ab titer (U/mL) 14-90 days post T/C in the observational cohort were retrospectively identified for further analysis (“flagged”). We then performed retesting of flagged blood samples with up-front 1:50 dilution, followed by serial 1:75 dilutions until two calculated results within expected assay variation (±10%) were returned. We defined prozone or “hook” effect if the Ab titer upon retesting was both ≥15% and ≥200 U/mL higher than original reported titer. Subsequently, all samples in the clinical trial were retested using the same protocol. In the observational cohort, we used a waiver of consent approved by IRB00248540. In the clinical trial, informed consent was obtained and approved by IRB00288774. Of 260 total samples from the observational cohort nine were flagged for further analysis of which all nine demonstrated a hook effect. Of 377 samples from the clinical trial (all rerun), 34/377 (9%) demonstrated a hook effect. Among the hook effect samples (n = 43), the original median (interquartile range; IQR) titer was 1950 (650, 4390) U/mL, and upon retesting this increased to 5685 (2981, 9853) U/mL representing a 1.6 (1.3-6.0)-fold increase (p = .03). Notably, a >2-fold rise in titer was seen in 18/43 (42%) participants (9 post-T/C, 9 post-vaccine). There was a marked hook effect (>700-fold increase on retest) present in samples from two participants who reported recent vaccination plus breakthrough infection (Figure 1). No qualitative change (i.e., negative to positive result or vice versa) was observed. Clinical characteristics of participants who provided samples that demonstrated a hook effect are shown in the Table 1. This study has limitations to consider. Participants in the clinical trial were SOTRs with known poor response to vaccination, and developed lower peak Ab titers, therefore, the reported hook effect proportion may be an underestimate as compared to the general vaccinated SOTR population. Additionally, due to resource availability, we did not rerun all observational cohort samples (post mAb), therefore the overall rate of hook effect in this group was not established; this, however, is likely a less clinically relevant use case than assessment of post-vaccine humoral response. In this study of SARS-CoV-2 Ab titers in SOTRs measured using a common clinical assay (Roche Elecsys), 100% of flagged mAb recipients and 9% of total vaccinee samples showed evidence of a hook effect. This was most prominent in those with highest analyte concentrations from two individuals with repeated antigen exposures in close temporal proximity (vaccination plus infection). We did not observe any change in qualitative test results. Providers, patients, and researchers utilizing anti-spike Ab testing as a component of risk stratification for SOTRs and other immunocompromised individuals at risk for poor SARS-CoV-2 vaccine response should be aware of this assay artifact. Up-front or targeted confirmatory dilution (e.g., following mAb or multiple antigen exposures) may be employed to identify hook effects and provide more accurate results for clinical and research purposes.5 Concept and design: All authors, Data analysis and interpretation: All authors, Manuscript writing: Aura T. Abedon, Teresa P.Y. Chiang, Andrew H. Karaba, Jennifer L. Alejo, William A. Werbel, Final approval of the manuscript: All authors, Accountable for all aspects of the work: All authors. The authors thank the participants of the Johns Hopkins COVID-19 Transplant Vaccine Study and the COVID-19 Protection After Transplant (CPAT) Consortium, without whom this research could not be possible. They also thank the members of the Johns Hopkins COVID-19 Transplant Vaccine study team, including Brian J. Boyarsky, MD PhD; Amy Chang, MD; Jonathan Mitchell, MBBS; Jake D. Kim, BS; Mayan S. Teles, BS; Divya D. Kalluri, BA; Michael T. Ou, MD; Ross S. Greenberg, MD; Jake A. Ruddy, MD; Alexa Jefferis BS; Michelle R. Krach, MS; Iulia Barbur, BSE; BS; Nicole Fortune Hernandez, BS; Letitia Thomas, Rivka Abedon, Alex Philip, Juhi Patel. They also thank Ms. Yolanda Eby for project support and guidance. This work was supported by generous donations from the Ben-Dov and Trokhan Patterson families, and by grants T32DK007713 (Dr. Alejo), K01DK101677 (Dr. Massie), K23DK115908 (Dr. Garonzik-Wang), K01DK114388 (Dr. Levan) from the National Institute of Diabetes and Digestive and Kidney Diseases, and grants K24AI144954 (Dr. Segev), 3U01AI138897-04S1 (Dr. Durand), K08AI156021 (Dr. Karaba) and K23AI157893 (Dr. Werbel) from the National Institute of Allergy and Infectious Diseases. AH Karaba has received consulting fees from Roche. DL Segev has the following financial disclosures: consulting and speaking honoraria from Sanofi, Novartis, CLS Behring, Jazz Pharmaceuticals, Veloxis, Mallinckrodt, Thermo Fisher Scientific, Regeneron, and AstraZeneca. CM Durand receives fees for serving on a grant review committee for Gilead Sciences. WA Werbel has received speaking honoraria for AstraZeneca, advisory board fees from Novavax, and consulting fees from GlobalData. The remaining authors of this manuscript have no financial disclosures or conflicts of interest to disclose as described by Clinical Transplantation. Data are not publicly available due to privacy restrictions. Reasonable requests for deidentified trials data may be made to the CPAT data coordinating center ([email protected]).