Food Allergies Can Persist After Myeloablative Hematopoietic Stem Cell Transplantation in DOCK8-Deficient Patients

To the Editor

DOCK8 deficiency is a highly morbid combined immunodeficiency that features recurrent sinopulmonary infections, viral skin infections, and severe food allergies.^1,2 Hematopoietic stem cell transplantation (HSCT) cures infection susceptibility in DOCK8 deficiency.^3–7 Whether HSCT also cures food allergy has not been systematically examined in humans.⁸ To gain insight into the etiopathogenesis of food allergy and its potential treatment, we studied food allergy in 12 DOCK8-deficient patients who underwent HSCT at the National Institutes of Health (NIH) Clinical Center.

First we retrospectively evaluated six patients who had received either matched related or unrelated donor cells following myeloablative conditioning (Patients 1–6 in Table 1) (see Methods in this article’s Online Repository).⁷ Of these, patients 2 and 4 reported histories of food-induced anaphylaxis before transplantation, and patient 5 reported a new food allergy after transplantation. Post-transplant, skin prick testing to eight common food allergens and specific IgE by ImmunoCAP confirmed sensitization to foods precipitating the reactions. Food challenges were not performed. A fourth patient (patient 1) reported oral pruritus to lentils before transplant and again when re-exposed more than one year after transplant. Donors in these cases were confirmed to have no history of food allergy.

Table 1.

Pre- and post- HSCT clinical food allergies, skin prick testing, and total serum IgE in the studied DOCK8-deficient patients.

	(Patient#) Age at HSCT	Months after HSCT at most recent allergy evaluation	Transplant Type^	GVHD prophylaxis type & duration of therapy after transplant	Gut GVHD	Clinical Food Allergies			Skin Prick Testing		Total Serum IgE (IU/mL)
						Foods causing reaction pre- HSCT	Foods causing reaction post- HSCT	Symptoms with most severe reaction pre-HSCT	Pre-HSCT Positive+	Post- HSCT Positive	Peak Pre- HSCT	Most Recent Post-HSCT
Patients studied retrospectively	1 18yoF	18	MRD	Cyclosporine: 12 months	Yes	Lentils	Lentils	Oral pruritus only	ND	ND	8,031	870
	2 10yoF	23	MRD	Tacrolimus: 8.5 months	No	Egg, Milk, Wheat, Soy, Sesame, Tree Nuts	Not Challenged	Anaphylaxis	ND	Egg, Milk, Wheat, Soy, Cashew (ND: Sesame)	6,690	1,058
	3 23yoM	21	MRD	Tacrolimus: 14 months	No	NONE	NONE	NONE	ND	ND	51,010	153
	4 27yoM	37	URD	Tacrolimus: 11 months	No	Peanut	Not Challenged	Anaphylaxis	ND	Peanut	1,162	17.4
	5 25yoF	24	URD	Tacrolimus: 3 weeks then, Cyclosporine: 14 months	No	NONE	Shrimp	Gastrointestinal symptoms only	ND	Shrimp	38.5	6.1
	6 16yoF	18	URD	Cyclosporine: 11 months	No	NONE	NONE	NONE	ND	ND	180	9.4
Patients studied prospectively	7 19yoM	12	URD	Tacrolimus: 9 months	No	Egg: resolved in early childhood, Tree Nuts persisted	Egg, No Tree Nut exposure	Anaphylaxis	-- Walnut Cashew	Egg Walnut --	6,398	1,224
	8* 21yoM	Died	URD	Tacrolimus: ongoing	Yes	Milk, Egg, Peanut	Not challenged	Anaphylaxis	Milk, Egg, Peanut, Shrimp	ND	31,403	627
	9 20yoF	12	haplo	Tacrolimus: 6 months	No	Egg, milk, wheat, peanut, cashew	Egg/milk, No exposures to others	Anaphylaxis	Egg, Wheat, Peanut, Cashew	Egg, Wheat, Peanut, Cashew	6,905	67.8
	10 13yoM	10	MRD	Tacrolimus: 7 months	No	Egg, Wheat, Peanut, Kiwi, Banana	Not challenged	Anaphylaxis	Egg, Wheat, Peanut	Egg, Wheat, Peanut	>6,000	915
	11 9yoF	6	URD	Tacrolimus: ongoing	Yes	NONE	NONE	NONE	NONE	NONE	2.0	2.9
	12 19yoF	3	haplo	Tacrolimus: ongoing	No	Kiwi, concentrated milk, concentrated egg	Kiwi, No exposure to others	Oral angioedema and pruritus	Milk, Egg, Cashew (Kiwi not done)	-- Egg, -- Kiwi	>6,000	594

^{^}MRD=Matched Related Donor, URD=Unrelated Donor, haplo=haploidentical

⁺Skin prick testing was performed for the following common food allergens: milk, egg, soy, wheat, peanut, walnut, cashew, shrimp for each patient pre- and post- HSCT unless otherwise indicated. All positive results are listed. − Negative result. ND: Not done.

^*Patient died prior to 3 months post-transplant; no follow up skin prick testing was performed.

Our observations of persisting or new food allergies were unexpected given anecdotal reports suggesting that HSCT cured food allergies, which could have reflected a selection bias.^4,5,9 Thus, we prospectively studied food allergies in the next six DOCK8-deficient patients transplanted at the NIH (Patients 7–12 in Table 1). This second group included two patients (patients 9 and 12) who had undergone related donor haploidentical transplants and had systemic allergic reactions to foods that they were already allergic to before transplant. Donors were confirmed to have no history of food allergy. Patient 9 was more than eight months after HSCT and more than two months off tacrolimus when she developed acute oral and facial angioedema, diffuse urticaria, vomiting, and difficulty breathing within minutes of eating oatmeal fortified with egg and milk. Her symptoms resolved after receiving epinephrine, diphenhydramine, and methylprednisolone. Prior to transplantation, she had had anaphylactic reactions to egg and milk. However, two months before her last reaction, her skin prick testing was positive to egg but not milk, suggesting egg as the culprit (see Table E1 in this article’s Online Repository). Similarly, patient 12 was 45 days out after HSCT when she developed oral and periorbital angioedema and diffuse urticaria and pruritus within 10 minutes of eating a kiwi fruit, with sensitization confirmed by skin prick testing (Table 1; and Table E1 in this article’s Online Repository). Years prior to transplant she too had had oral pruritus and lip angioedema after eating kiwi and had subsequently avoided it entirely.

Among the patients studied prospectively, a third patient who underwent matched unrelated donor transplantation reported that a previously resolved food allergy had returned. Patient 7 was three months out of HSCT when he developed cramping abdominal pain, vomiting, diarrhea, and headache within 15 minutes of eating scrambled eggs. His symptoms occurred on two more occasions, but never when he ate baked eggs. He had had similar symptoms in his early school age years but had been eating eggs freely for the decade prior to transplant. Skin prick testing confirmed that he had acquired new reactivity to egg after transplantation (Table 1; and see Table E1 in this article’s Online Repository). The donor was confirmed to have no history of food allergy. Patient 7 also had a history of anaphylaxis to walnut as recently as three years prior to HSCT. Because of persisting positive skin prick testing, he continued to strictly avoid tree nuts after transplantation.

We observed that total serum IgE levels plummeted after HSCT but remained high in most patients (Figure 1). Food-specific IgE levels also remained high for months after transplant in several patients, even at levels with >95% positive predictive value for clinical food reactivity for some (see Table E2 in this article’s Online Repository). Moreover, skin prick testing revealed persisting mast cell reactivity to food allergens, which seemed to correlate best with the anaphylactic episode described post-transplant in patient 9 (see Table E1 in this article’s Online Repository). Persistence of allergen sensitization occurred regardless of donor type. The possibility of long-lived, host-derived, IgE-producing plasma cells in the bone marrow could explain allergy persistence. Indeed, we observed that bone marrow chimerism approached but did not reach 100% (see Table E3 in this article’s Online Repository).

Figure 1.

Total serum IgE in DOCK8-deficient patients after HSCT. Symbols used to represent patients: patient 1, blue-filled circle; patient 2, red-filled square; patient 3, green-filled triangle; patient 4, purple-filled triangle; patient 5, orange-filled diamond; patient 6, black-filled circle; patient 7, brown-filled square; patient 8, open yellow circle; patient 9, navy-filled triangle; patient 10, burgundy-filled diamond; patient 11, open blue circle; patient 12, green asterisk.

Although HSCT in DOCK8-deficient patients is now considered standard of care, its outcome for food allergy remains less clear. In two DOCK8-deficient patients who had full donor chimerism after HSCT, food allergies were only mentioned as having resolved.^4,5 In another DOCK8-deficient patient with full donor chimerism after HSCT, multiple undescribed food allergies resolved while lentil allergy (the only one to cause anaphylaxis in this patient) persisted and total IgE levels remained abnormally high.⁶ In a fourth patient who had mixed donor peripheral blood chimerism (53% mononuclear cells, 6% granulocytes, 98% T-cells, 35% B-cells), multiple food allergies persisted although they were less severe.³ By contrast, all of our patients who had food allergy reactions after transplantation had 100% peripheral blood donor chimerism (see Table E3 in this article’s Online Repository).

Our study is limited in that we did not perform double blind placebo-controlled food challenges before and after HSCT. However, the two systemic reactions (patients 9 and 12) and one local reaction (patient 1) to foods that had previously caused reactions prior to transplant strongly support persistence of food allergy in these three patients. Tacrolimus has been associated with new-onset food allergy after solid organ transplantation.¹⁰ Although its use in our patient cohort could have contributed to some of the allergy we observed, this seemed unlikely in two patients. Patient 1 received only cyclosporine for graft versus host disease (GVHD) prophylaxis. Patient 5 received tacrolimus for only three weeks before it was switched to cyclosporine. Furthermore, in no previous reports of persisting allergies did DOCK8-deficient patients receive tacrolimus^3,6.

Our case series describes what is to our knowledge the first systematic study of food allergy in HSCT, which may have implications for understanding food allergy more broadly in patients who do not have DOCK8 deficiency. We have observed that food allergy is not always cured after HSCT, at least not initially, even when 100% peripheral blood donor chimerism is achieved. Thus, strict food avoidance diets should continue to be followed until appropriate allergy testing and medically supervised food challenges can be performed. The overall trends toward decreasing total IgE, decreasing food-specific IgE, and decreasing skin wheal sizes, together suggest that these patients may have an increased probability to eventually outgrow their food allergies than their food allergic peers who have not been transplanted. Long-term follow up of these patients will be informative.

Corinne S. Happel, MD

Kelly D. Stone, MD, PhD

Alexandra F. Freeman, MD

Nirali N. Shah, MD

Angela Wang, RN

Jonathan J. Lyons, MD

Pamela A. Guerrerio, MD, PhD

Dennis D. Hickstein, MD

Helen C. Su, MD, PhD

Abbreviations used

CGH

comparative genomic hybridization

F

female

GVHD

graft versus host disease

Haplo

haploidentical donor

HSCT

hematopoietic stem cell transplantation

IgE

immunoglobulin E

M

male

MRD

matched related donor

ND

not done

NIH

National Institutes of Health

URD

unrelated donor

yo

year old