Evaluation of solid-phase panreactivity with negative direct antiglobulin testing

Twenty consecutive patients with panreactive NSP antibodies were identified between November 2022 and May 2023. In addition to their panagglutinins, these patients were chosen for study because they all had (1) a negative PEG-tube antibody detection test, (2) a negative PEG-tube autocontrol, and (3) a negative DAT. A summation of patient demographic data is found in Table 1. Half of the patients were pregnant women, which resulted in a lower mean age of the female study patients. Although it appears that both women, in general, and pregnant women, in particular, are overrepresented in the study, these demographic features are representative of the total distribution of patient types evaluated by the laboratory (data not shown).

A summation of the serologic data evaluated in the study is found in Table 2. The overall, predominant strength of SPRCA panreactivity was evenly distributed across reaction grades (1+ to 3+). Eluate testing revealed that 55 percent of the eluates tested by PEG-tube were panreactive and consistent with warm-reactive autoantibodies. However, 85 percent of eluates tested in the SPRCA environment were panreactive and consistent with warm-reactive autoantibodies. Six discrepant cases were associated with relatively weak SPRCA plasma reactivity (1+). In these cases, PEG-tube eluate testing was negative and SPRCA eluate testing demonstrated a panagglutinin. In addition, the strength of reactivity of the eluates tested by SPRCA were invariably more strongly reactive (1+ to 2+ greater strength of reactivity) than eluates tested by PEG-tube (data not shown). Interestingly, there were three eluates that were nonreactive by both PEG-tube and SPRCA. All three had strongly reactive SPRCA plasma reactivity (3+).

Finally, each study patient’s electronic medical record was interrogated for evidence of hemolysis by reviewing narrative reports (e.g., consultation and progress notes) and laboratory results (Hb/Hct, reticulocyte count, bilirubin, haptoglobin, and LDH values). There was no clinical or laboratory evidence of hemolysis. Although there were gaps in the laboratory values obtained for each patient, 16 of 20 had normal or near-normal Hb and Hct values. The remaining four patients were anemic (Hct values in the 20s; normal range 36–46% for adult women and 41–53% for adult men), but each patient had evidence of active bleeding. Thus, although the vast majority of SPRCA panagglutinins represent warm-reactive autoantibodies, none were associated with overt hemolysis (i.e., warm autoimmune hemolytic anemia).

SPRCA assays have a long history of use as a primary platform for blood group antibody detection testing because of their sensitivity and conduciveness to automated testing methods.³ However, sensitive antibody identification must be balanced against the detection of clinically insignificant antibodies and nonspecific reactions. Several publications have acknowledged that, although SPRCA methods are quite sensitive, they are also prone to a higher rate of nonspecific reactions than other platforms.^4–6 One pattern of such reactivity is panagglutination, although the exact mechanism for this phenomenon remains unclear.^3,7,8 In addition, SPRCA assays appear to be more sensitive for the detection of warm-reactive autoantibodies than other primary antibody detection methods.^4,6 In a previous study of NSP antibodies, the authors identified an interesting pattern of reactivity: panreactive SPRCA plasma testing with a negative PEG-tube test and a negative DAT.² A preliminary study indicated that at least a portion of these antibodies actually represented warm-reactive autoantibodies.² In the current study, we extend those findings by studying a greater number of cases correlated with both patient clinical characteristics and more expansive serologic findings.

The most notable finding in our study is that even with a negative DAT, the vast majority of SPRCA panagglutinins are warm-reactive autoantibodies (based on a panreactive eluate). This finding is surprising, but perhaps should not be given the fact that (1) SPRCA testing is sensitive for warm-reactive autoantibodies and (2) relatively speaking, SPRCA plasma testing may be more sensitive than a standard gel- or tube-based DAT. Fortunately, these particular “SPRCA-only warm-reactive autoantibodies” do not appear to be associated with clinical hemolysis. This observation likely results from there being too little immunoglobulin bound to the patient’s own RBCs (i.e., a negative DAT) to result in overt extravascular hemolysis. Finally, panagglutinating antibodies of this type are easily managed through use of an alternative antibody detection method. Specifically, these antibodies had completely negative antibody detection tests by PEG-tube, making the detection of underlying alloantibodies straightforward.

A final and interesting serologic finding identified in this study is seen in the three case subjects who had nonreactive eluate testing in SPRCA and PEG-tube testing environments. Two of these three patients were pregnant, and the third patient had active bleeding (hematuria). All, however, had 3+ plasma testing using SPRCA. We can only speculate as to what these panagglutinins represent, but the most likely possibilities include (1) a warm-reactive autoantibody despite negative eluate testing; (2) an antibody to a high-prevalence blood group antigen, including high-titer, low-avidity reactivity, that is only demonstrable in SPRCA; and (3) an antibody or other agglutinin that binds a component of the SPRCA testing matrix (e.g., a component of the solubilized internal RBC membrane).⁹ Of these, the second and third possibilities seem most likely.

Like all studies, our study has strengths and weaknesses that must be acknowledged. Because our sample size is relatively small (n = 20), corroborative studies should be performed to verify our findings, but it is difficult to believe that additional samples would lead to radically different results, especially since the current results are consistent with results obtained in a pilot study.² Another potential weakness is that the patients selected for study, although consecutive in nature, represent a convenience sample of those case subjects who presented to the laboratory in the 7-month time frame of the study. It could be that other patients chosen at other times might have yielded different results. Another limitation to the study is the fact that DATs were performed using gel testing instead of tube methods. This difference in testing adds another variable when comparing PEG-tube antibody detection test results with SPRCA results. Some of these patients may have had a positive DAT using a tube method. Finally, it would have been interesting to follow the 10 pregnant study patients to determine how these panagglutinins affected their neonates, once delivered. It seems unlikely that neonates would be at increased risk for hemolysis or even a positive DAT, particularly because their mothers were unaffected, but this analysis was beyond the scope of the current study.

In conclusion, nonspecific serologic reactivity appears to be more common in SPRCA testing than other antibody detection platforms. This finding is likely due to the enhanced sensitivity of SPRCA for IgG antibodies. A distinct SPRCA-only plasma reactivity pattern is panagglutination. Despite negative PEG-tube testing and DAT results, the majority of these antibodies appear to be warm-reactive autoantibodies. Fortunately, these “interfering” panagglutinins do not appear to be clinically significant, with respect to hemolysis, and they are easily managed by turning to an alternative testing method such as PEG-tube testing. Further investigation is warranted to better understand the clinical and serologic features of nonspecific SPRCA antibodies, particularly antibodies that are ultimately determined to be warm-reactive autoantibodies.