Q&A: Testing for polyethylene glycol in drug allergies remains challenging

Key takeaways:

• Skin prick and intradermal testing do not truly confirm polyethylene glycol allergy.
• Patients may experience multiple episodes of anaphylaxis without a clear trigger.
• Larger, multicenter cohorts are needed.

Hypersensitivity to polyethylene glycol may be rare, but when reactions to PEG in drugs do occur, they often are due to molecular weights of 3350 and higher, according to a study published in Annals of Allergy, Asthma & Immunology.

A better understanding of how these weights trigger reactions could improve outcomes among patients who rely on drugs that include PEG for care as well as among patients who use vaccines such as mRNA COVID-19 vaccines that may include PEG.

Healio spoke with John Accarino, MD, a physician in clinical allergy/immunology at Massachusetts General Hospital, who was not one of the study’s authors, for an outside perspective of these findings and what they may mean for outcomes.

Healio: Do you think these findings are surprising or significant?

Accarino: These findings are very significant, as it takes a bit of effort to fully characterize even one patient for PEG allergy. Even with a multicenter retrospective study over 7 years, these researchers were able to include only 44 patients. These numbers are the largest we can expect from studies focusing on less common specific drug allergies like PEG.

PEG testing is performed regularly in our clinic. PEG evaluations were more common in recent years due to the previous concerns of PEG as a potential culprit for reactions to mRNA COVID-19 vaccines. This concern has been alleviated by subsequent studies showing no association as well as evidence of true PEG-allergic patients tolerating mRNA COVID-19 vaccines without reaction.

The authors’ study is not confounded by those evaluations because they did not include patients with evaluations after December 2020.

Even after removing the COVID-19 vaccine PEG patients (most of whom had negative testing), it was surprising to me to see that 42 cases were confirmed out of 44 patients evaluated (95%). This is likely related to the inclusion of patients with a high clinician-determined pretest probability of PEG excipient allergy. Forty-one of 42 patients with data from the index reaction had Grade II or Grade III Brown reactions, which is a higher percentage of severe reactions compared with many of our PEG evaluations.

The impact of these data would make me interested in going back and measuring the specific rate of confirmation for our own PEG evaluative cohort over the last few years.

Using penicillin allergy as a comparison, we are finding that even the highest-risk histories for suspicion for a penicillin allergy are only about 50% predictive of confirmation, with very high-risk histories only leading to 20% confirmed positivity (using PEN-FAST scores of 3, 4 and 5).

Another impressive aspect of this study is the availability of acute mast cell tryptase (MCT) in 19 of its patients, which we usually do not have for most patients. The rise in tryptase for all 19 patients averaging 53 µg/L is striking. I was very surprised that the range of change in MCT was 9 µg/L to 109.7 µg/L. This means that every patient who had both baseline and acute tryptase measured had, at minimum, an increase in their tryptase by 9 µg/L.

This fact is surprising and significant to me for clinical practice as a predictive marker, because I do not usually rely on tryptase as heavily compared with reported features of the reaction history. This is because people can still have confirmed drug allergy reactions even without an acute rise in tryptase levels. I would have expected that at least some of these confirmed patients to have had normal acute tryptase levels, which was not the case.

Finally, positive skin prick testing does not truly “confirm” allergy, although it leads us to keep allergy labels and recommend continued avoidance. While 26 of these patients were skin prick test (SPT) positive and 15 were positive in intradermal testing (IDT) and subsequently labeled as “confirmed PEG allergic,” the only gold standard currently is a medication challenge. We do not have controlled studies proving that all 26 of those patients would truly react to PEG challenge.

This is a difficult hurdle in the field since it takes immense effort and extensive risk-benefit conversations to conduct a study where SPT-positive patients may receive an observed drug challenge that would further validate the positive predictive value of SPT. So, it is a nuanced point where they should avoid the drugs due to the predictive values of the procedures that we currently have available, but we do not truly know how they would do after a repeat exposure.

Healio: How else do these findings compare with your experience?

Accarino: In our experience, I completely agree that diagnosis is often delayed and often requires multiple reactions to multiple medications before a clear picture begins to emerge. Across thousands of drug allergy evaluations, excipients only account for a handful of reactions. With the first medication reaction, we would expect the active ingredient to be the culprit. With the second or third reaction, it still may be difficult to differentiate if this is a true underlying excipient allergy or if this is a mimicking diagnosis.

Certain drugs, such as injectable steroids, acting as the culprit for the index reaction would raise suspicion much earlier than other drugs. Yet, even in injectable steroids, hypersensitivity to other excipients such as carboxymethylcellulose may be alternative causes of reactions.

One exception would be an immediate reaction to a PEG-only product, such as macrogol laxatives, where PEG is the primary ingredient. These would highly suggest PEG allergy and seem to be the most common culprits, as outlined in the paper. In the most recent 2022 Drug Allergy Practice Parameter Update, there is a conditional recommendation based on low certainty of evidence to consider excipient hypersensitivity “in patients with a history of anaphylaxis to 2 structurally unrelated drugs or products that share a common excipient, (e.g., injectable corticosteroids; PEG-based laxatives).”

Patients who experience chronic spontaneous urticaria due to intrinsic mast cell triggers such as heat, exercise, irritation, or viruses will frequently search for extrinsic causes. Medications taken around the time of chronic hive eruptions are frequently blamed, and these patients may end up having allergy lists with many medications incorrectly labeled as allergy that were never confirmed. This may lead to the concern that an inactive ingredient such as PEG is the root cause without exhibiting associated evidence via skin testing or challenge.

This highlights the need for objective evaluation with procedures such as SPT, IDT and challenges to sparse out these confounders.

True PEG allergy can have very significant impacts on one’s life and can severely limit one’s options in health care. These patients may have multiple episodes of anaphylaxis due to various preparations of common medications and medical supplies that may contain PEG.

Healio: How can doctors use these findings to improve care?

Accarino: Doctors can use these findings to be on high alert for PEG allergy even beyond the classic culprits of bowel preparations and injectable steroids.

The 10% of patients who reacted due to PEG in oral penicillin V, 3% for amoxicillin, 1% for cefuroxime, 1% for flucloxacillin, and all the other drugs included in the study also push me to keep PEG in the back of my mind for antibiotic allergy evaluations, which are the most common presenting classes for drug allergy evaluation.

Furthermore, for any patient who had negative SPT to these antibiotics but then failed a subsequent direct challenge (in very rare cases), PEG may be a possible next step to better characterize the etiology beyond just avoidance of that antibiotic.

Lastly, the fact that all 19 patients measured had a significant rise from baseline to acute tryptase is very helpful to support utility as a predictive marker in the future.

Healio: What should the next step be in this research?

Accarino: One next step would be to perform these types of studies in larger multi-site prospective cohorts. Drug allergy is a heterogenous field, and specific reactions, presentations, and medications may have lower or higher respective rates of confirmation. We do not know if PEG is uncommonly diagnosed due to a low incidence in the population or if it is currently underdiagnosed due to clinicians having a low index of suspicion. Surveys of provider awareness about PEG allergy would be helpful to measure this.

There also is a need for more translational research to assess specific anti-PEG antibodies and PEG-associated immune responses. In IgE-mediated food allergy or environmental allergy, there is usually a larger protein antigen culprit. But in drug allergy, the molecules are so small that there are many other mechanisms we must consider. There may also be epitopes that exist for longer chains of high molecular weight PEG that are not present in shorter chains of lower molecular weight PEG. These types of basic science studies are direly needed to better understand the pathways involved.

As discussed in the paper, there are no commercially available SPT reagents used for PEG evaluation. Even in this paper, some centers used PEG 3350, and others used PEG 4000.

It seems that some patients are specifically sensitive to only high molecular weight PEG, but there are also patients who experience symptoms with both high and low molecular weight PEG. There is some evidence of this variability in this paper, and it needs to be further explored in larger studies. As the article states, patients are often advised to avoid all PEG-containing drugs, which we may find to be overly cautious in a subset of patients.

References:

• Hagedoorn NN, et al. PLOS Med. 2020.;doi:10.1371/journal.pmed.1003208.
• Kayode OS, et al. Ann Allergy Asthma Immunol. 2024;doi:10.1016/j.anai.2024.03.022.
• Khan DA, et al. J Allergy Clin Immunol. 2022;doi: 10.1016/j.jaci.2022.08.028.
• Trubiano JA, et al. JAMA Intern Med. 2020;doi:10.1001/jamainternmed.2020.0403.

For more information:

John Accarino, MD, can be reached at jaccarino@mgh.harvard.edu.