Direct oral challenges rival standard of care in assessing low

Copaescu AM, et al. JAMA Intern Med. 2023;doi:10.1001/jamainternmed.2023.2986.

Copaescu AM, et al. JAMA Intern Med. 2023;doi:10.1001/jamainternmed.2023.2986.

Direct oral challenges in adults at low risk for penicillin allergy were safe and not inferior to the standard-of-care oral challenge following two-stage skin testing, according to a study published in JAMA Internal Medicine.

This procedure may facilitate the removal of a larger number of penicillin allergy labels, Elizabeth Phillips, MD, the John A. Oates Chair in Clinical Research at Vanderbilt University Medical Center, and colleagues wrote.

“The study was conceived out of an unmet need to have higher-level evidence to support removing a label of penicillin allergy in low-risk patients, and especially low-risk adults where the level of evidence to support direct oral challenge without preceding skin testing is weakest, without the need for specialized skin testing,” Phillips told Healio.

“Getting to an allergist and immunologist for specialized skin testing is a limited resource,” she continued. “When we surveyed infectious disease physicians on this 7 years ago, almost one in four said they had no access to antibiotic allergy testing services.”

Conducted across six clinics in the United States and Australia, the PALACE study comprised 377 adults (median age, 51 years; 65.5% women) with a PEN-FAST score of less than 3, indicating low risk for penicillin allergy.

One of the 187 patients (0.5%) in the intervention group, only receiving a direct oral challenge, and one of the 190 patients (0.5%) in the control group, receiving the standard of care, had positive oral penicillin challenges consistent with immune-mediated reactions, for a risk difference between the groups of 0.0084 percentage points (95% CI, –1.22 to 1.24). This was below the noninferiority margin of 5 percentage points, with a risk ratio of 1.02 (90% CI, 0.1-10.34).

The two patients with positive results had mild cutaneous skin reactions including immediate diffuse rash or urticaria that resolved with a single dose of antihistamines.

There were 22 cumulative adverse events among 20 patients in the intervention group and 24 adverse events among 21 patients in the control group in the 5 days after the oral penicillin challenge for a risk difference of –0.36 percentage points (95% CI, –6.64 to 5.93) and risk ratio of 0.97 (95% CI, 0.54-1.73).

Also, the intervention group experienced nine immune-mediated adverse events in the following 5 days, a comparable figure to the 10 experienced in the control group (risk difference, –0.45 percentage points; 95% CI, –4.87 to 3.96). These adverse events occurred after a median of 4 hours (interquartile range [IQR], 0.67-16.67) in the intervention group and 6 hours (IQR, 2.09-35.1) in the control group.

Reactions occurring more than an hour after the oral penicillin challenge included delayed diffuse rash or urticaria in six patients in the intervention group (3.2%) and three patients in the control group (1.6%) for a risk difference of 1.63 percentage points (95% CI, –1.46 to 4.72), with no serious adverse reactions.

Nine of the events in the intervention group and four in the control group required treatment, but none of them required hospitalization or presentation at the ED.

The study removed penicillin allergy labels for 186 of the 187 patients (99.5%) in the intervention group and 186 of 190 patients (97.9%) in the control group, with a risk difference of 1.57 percentage points (95% CI, –0.72 to 3.86).

Due to positive intradermal skin tests, four patients in the control group were excluded from the oral challenge, which explains the difference in the efficacy of de-labeling in the intervention group, Phillips and colleagues wrote.

Based on these findings, the researchers called direct oral challenge a safe and effective method for assessing patients with low-risk penicillin allergies in a predominantly white, adult, outpatient population.

“I think the study confirmed the hypothesis that in low-risk patients with a penicillin allergy label, going straight to an oral challenge is no worse than using the standard of penicillin skin testing followed by the oral challenge with a penicillin,” Phillips said.

Although the number of participants was relatively small, Phillips noted, the study also suggested that patients at very low risk may see false-positive skin tests, leading to fewer low-risk patients among those who get skin tests who are de-labeled.

“This just follows the basic principles of epidemiology, where a false-positive test is increased in those where the pre-test probability is very low,” Phillips said. “This suggests that in low-risk patients where the pre-test probability of a true penicillin allergy is low, the responsible approach would be to go directly to challenge with a penicillin.”

The researchers also cautioned that 94% of the participants in the study had PEN-FAST scores of 0 or 1, limiting the generalizability of the study to patients with scores of less than 2.

Still, the researchers continued, direct oral challenges require fewer resources and less time than skin testing and are less expensive as well. Additionally, these challenges can be performed outside of specialist allergy settings and offer a scalable approach for addressing low-risk, unverified penicillin allergies in diverse settings.

“These findings are highly significant because they mark the beginning of a new era where we can truly scale up the removal of a false penicillin allergy label that many patients have carried for decades,” Phillips said. “This study will empower different types of physicians with a readily available tool to do this.”

In fact, Phillips said, all physicians should use PEN-FAST to identify patients at low risk for penicillin allergy despite their allergy label and then de-label them in their office using a single dose of an oral penicillin.

“This will also identify patients who have nonallergic histories such as nausea or headache for which the penicillin allergy label could be removed by history alone,” Phillips said.

Further, Phillips said, the study reinforces the use of appropriate and first-line antibiotics such as penicillin, with significant consequences for the patient as well as for public health.

“Broad-spectrum antibiotics may increase the risk of antibiotic resistance and superbugs, serious infections such as Clostridioides difficile and emerging infectious disease threats such as Candida auris,” Phillips said.

Next, Phillips said, it will be key to investigate the longer-term outcomes of penicillin allergy de-labeling as well as what needs to be done to support and reinforce it.

“We need to make sure when we de-label a patient of a penicillin allergy that this is a permanent process and that patients are not at risk to be re-labeled,” she said.

Health care in the United States is quite fragmented, Phillips continued, adding that once patients are de-labeled, that information needs to get to their pharmacy and other providers.

Also, she said, 75% of patients with penicillin allergy labels received those labels by age 3 years, generally because penicillin allergy was confused with a viral rash. Phillips called for greater awareness of these errors and efforts in preventing penicillin allergy labels by recognizing patients at low probability of having a true allergy based on history.

“In addition, since the adults we see have been labeled sometimes for more than 50 years, we need to understand about the psychology of a penicillin allergy label and the anxiety and identity associated with that, and how this may play into personalized approaches to penicillin allergy de-labeling,” she said.

Additionally, Phillips called for better approaches in de-labeling patients from underrepresented groups and uninsured populations, including those at high risk for morbidity and mortality from infections, such as pregnant women and their unborn children as well as those who are having surgery.

“Programs set up in the inpatient setting such as a predischarge penicillin de-labeling program based on oral challenge with a penicillin prior to discharge, akin to vaccination, could minimize disparities in care,” Phillips said.

Elizabeth Phillips, MD, can be reached at [email protected].

Monica Kraft, MD

Although not particularly surprising, it is significant that this study demonstrated clearly that penicillin skin testing did not add significant additional diagnostic value for patients whose reported allergy history was low risk.

We have known for years that the vast majority of people who report a penicillin allergy are not allergic, whether due to loss of that sensitization over the years since the reaction or misattribution of other symptoms as an allergic reaction. Most oral challenges to penicillin that I perform in clinic are negative and confirm the patient is not allergic.

Consideration for direct oral challenge without prior skin testing for patients with low-risk history was included in the 2022 drug allergy parameters update, so this has become my general practice as well. It is helpful to have this study demonstrate in a prospective, randomized way this is a safe and reasonable approach.

Eliminating the need for skin testing, which requires specialized extracts and personnel trained in placing and interpreting the test, can increase access for patients interested in clarifying their penicillin allergy.

This study included low-risk penicillin reaction histories only, so for higher-risk patients, skin testing is likely to remain necessary. Future studies should help discern which patients with an intermediate-risk reaction history would most benefit from skin testing prior to oral challenge.

Joanna L. Stollings, PharmD

I think these findings are very significant and indicate that oral challenge is not inferior to skin testing/oral challenge. These findings also are very consistent with what I have seen. I have personally given more than 300 patients a low-dose oral challenge (250 mg per oral amoxicillin) without skin testing based on a very similar algorithm to PEN-FAST for inclusion with very minimal reactions.

Doctors or other providers such as pharmacists can apply this practice to their daily routine. By using a similar algorithm, patients can be easily identified and quickly given an oral dose of penicillin.

The next step in this research should be testing this protocol in a pragmatic fashion to see if it can be implemented in an entire hospital or in hospital systems successfully with the same results.