Pneumocystis jirovecii Pneumonia Prophylaxis with Intravenous Pentamidine in Adult Allogeneic Hematopoietic Stem Cell Transplant Patients

ABSTRACT

The purpose of this single-center retrospective case series was to evaluate the efficacy and safety of 300-mg once-monthly intravenous (IV) pentamidine prophylaxis in 702 adult allogeneic hematopoietic stem cell transplant (HSCT) patients. We observed no cases of Pneumocystis jirovecii pneumonia (PJP) following IV pentamidine administration. Breakthrough Nocardia and Toxoplasma infections were observed in 7 (1%) and 5 (0.7%) patients, respectively. The most commonly reported adverse event was nausea. Monthly IV pentamidine is a reasonable alternative to trimethoprim-sulfamethoxazole (TMP-SMX).

INTRODUCTION

Pneumocystis jirovecii pneumonia (PJP) is an infection of concern for hematopoietic stem cell transplant (HSCT) recipients (1–3). Without prophylaxis, the rate of P. jirovecii infection ranges from 5 to 16% among allogeneic HSCT patients, with mortality rates ranging from 30 to 60% (4–6). Prophylaxis is recommended for at least 6 months following transplant or longer with continued immunosuppressants (1, 2, 7). With appropriate prophylaxis, PJP incidence decreases to 1 to 6% (4).

Trimethoprim-sulfamethoxazole (TMP-SMX) is the preferred prophylactic agent, given robust data and superior efficacy, compared to alternatives (1, 2, 7). TMP-SMX also offers broad prophylactic coverage of other pathogens (2, 8). However, there is concern regarding myelosuppressive properties of TMP-SMX and resultant delayed engraftment in the HSCT population (9).

Once-monthly intravenous (IV) pentamidine has been described as PJP prophylaxis for patients with human immunodeficiency virus (HIV) infection; however, data are limited in the HSCT population. Use in HSCT patients was described in a combined autologous and allogeneic HSCT pediatric population (10–12). Small retrospective studies detailed successful PJP prophylaxis for adults using IV pentamidine following allogeneic HSCT (13, 14). These studies support the use of IV pentamidine, although they are limited by the heterogeneity of the populations examined.

At our institution, patients receive TMP-SMX from the first day of their conditioning regimen through the evening 2 days prior to transplant (day −2) (15). Patients receive a 300-mg dose of IV pentamidine by day +25 during the initial posttransplant period, with repeat administration every 28 days until the patients are able to tolerate oral TMP-SMX. IV pentamidine is our standard second-line prophylaxis for patients who are unable to achieve sustained platelet counts or have contraindications to TMP-SMX. The objective of this study was to evaluate the safety and efficacy of IV pentamidine for PJP prophylaxis.

RESULTS

In total, 702 adult patients received at least 1 dose of IV pentamidine. No breakthrough PJP cases were identified. Breakthrough Nocardia and Toxoplasma infections were observed in 7 (1%) and 5 (0.7%) patients, respectively. One patient was treated empirically for listeriosis but not definitively diagnosed.

Of the 702 patients, 280 patients underwent full chart review. Complete data extraction was limited due to time constraints. The 280-patient sample included 3 patients with a history of PJP or PJP diagnosis posttransplant. The remainder of the patients were randomly chosen. Patient demographic features are provided in Table 1. Of note, 219 (78.2%) patients developed some degree of graft-versus-host disease (GVHD). Most patients were also receiving concomitant immunosuppression.

TABLE 1.

Patient demographic characteristics (n = 280)

Parametera	Finding
Age (median [range]) (yr)	55 (19–76)
Gender (no. [%])
Female	106 (38)
Male	174 (62)
Cancer diagnosis (no. [%])
Acute leukemia (ALL, AML, or APL)	154 (55)
Chronic leukemia (CLL, CML, or CMML)	24 (8.6)
Lymphoma	34 (12.1)
Myelodysplastic syndrome	50 (17.8)
Multiple myeloma	12 (4.3)
Otherb	6 (2.2)
Stem cell donor type (no. [%])
Matched related donor	100 (35.7)
Matched unrelated donor	164 (58.6)
Mismatched unrelated donor	16 (5.7)
Conditioning regimen (no. [%])
BuCy/TBuCy	88 (31.4)
BuFlu with or without TBI	66 (23.6)
CyTBI	38 (13.5)
FluMel with or without TBI	52 (18.6)
BEAM	15 (5.4)
Otherc	21 (7.5)
Other medicationsd (no. [%])
Corticosteroidse	226 (80.7)
Cyclosporine	165 (58.9)
Methotrexate	29 (10.4)
Mycophenolate	59 (21.1)
Sirolimus	18 (6.4)
Tacrolimus	137 (48.9)
Ganciclovir/valganciclovir	118 (42.1)
Toxoplasma IgG antibody positivef (no. [%])	34 (12)
History of P. jirovecii infection (no. [%])	2 (0.7)
Developed GVHD (no. [%])	219 (78.2)

^aALL, acute lymphocytic leukemia; AML, acute myeloid leukemia; APL, acute promyelocytic leukemia; CLL, chronic lymphocytic leukemia; CML, chronic myeloid leukemia; CMML, chronic myelomonocytic leukemia; TBI, total-body irradiation; BuCy, busulfan plus cyclophosphamide; TBuCy, TBI plus BuCy; BuFlu, busulfan plus fludarabine; CyTBI, cyclophosphamide plus TBI; FluMel, fludarabine plus melphalan; BEAM, regimen consisting of carmustine, etoposide, cytarabine, and melphalan.

^bOther cancer diagnoses included aplastic anemia and myelofibrosis.

^cOther conditioning regimens included BuFlu plus rituximab (Rituxan), cyclophosphamide plus anti-thymocyte globulin (ATG) (Atgam), cyclophosphamide plus fludarabine plus TBI, fludarabine plus cyclophosphamide plus ATG plus TBI, FluMel plus ATG, TBI, treosulfan plus FluMel, treosulfan plus fludarabine plus TBI.

^dOther medications included other potentially myelosuppressive medications, including posttransplant GVHD prophylaxis medications (e.g., calcineurin inhibitors, mycophenolate, or methotrexate), given concomitantly with pentamidine.

^eCorticosteroids includes all patients who received high-dose corticosteroids concomitant with pentamidine treatment. High dose was defined as doses equivalent to 20 mg prednisone daily consecutively for ≥4 weeks.

^fPrior to transplant.

A total of 1,087 doses of pentamidine were administered to the 280 patients, with a median of 3 pentamidine doses per patient (range, 1 to 46 doses). Alternative prophylaxis was initiated for 43 patients posttransplant but later switched to pentamidine; 10.7% (n = 30) initially started TMP-SMX posttransplant, and 4.6% (n = 13) were initiated with dapsone. Subsequent changes between different PJP prophylaxis regimens occurred for 197 patients; in some cases, multiple changes occurred. There were 118 patients who received consecutive doses of pentamidine and never switched to TMP-SMX prophylaxis. The rationales for continuing pentamidine administration included persistent thrombocytopenia or cytopenia (n = 93), TMP-SMX allergy or intolerance (n = 14), patient preference (n = 1), and other clinical considerations (n = 10). Thirteen patients died after receiving the initial dose of pentamidine and before initiating TMP-SMX. None of those deaths was attributed to P. jirovecii infection. One-tenth of the patients (n = 29) missed at least one dose, and 17.8% of the patients (n = 50) received at least one late dose.

The most commonly reported adverse event was nausea (n = 47 [16.8%]), followed by generalized pain or discomfort during infusion (n = 13 [4.6%]). Other adverse effects reported less frequently included indigestion, diarrhea, numbness, hypotension, rash, dizziness, tachycardia, hypomagnesemia, and weakness (n ≤ 4). There were no instances of acute kidney injury attributed to pentamidine infusions.

DISCUSSION

Our 702-patient study adds to the data supporting the use of IV pentamidine as PJP prophylaxis for adult allogeneic HSCT patients. No breakthrough PJP cases were identified during IV pentamidine prophylaxis. Of note, 1 patient in our study was diagnosed with PJP by positive PCR and radiographic findings 15 months following transplant. This diagnosis did not meet the definition of breakthrough infection while receiving IV pentamidine because the patient was diagnosed 97 days after the last dose and had been transitioned to TMP-SMX. It is possible that the colonization or infection developed during the pentamidine course.

Of the 702 study patients, 7 (1%) developed infections due to Nocardia species and 5 (0.7%) due to Toxoplasma. These data are largely lacking in previous studies, with only one study reporting rates of Toxoplasma breakthrough (10). In our study, the breakthrough rates observed for both Nocardia species and Toxoplasma were lower than those reported for patients receiving TMP-SMX prophylaxis, suggesting that HSCT patients receiving IV pentamidine were not at increased risk of these infections; however, further exploration is warranted (7, 16, 17).

Our study is limited by the retrospective design and loss to follow-up monitoring. Data collection for HSCTs performed between 2007 and 2008 was also limited, because these occurred prior to implementation of the current electronic health record (EHR). Only 280 of 702 eligible patients underwent full chart review, although all 702 were evaluated for evidence of PJP and Listeria, Nocardia, and Toxoplasma infections during prophylaxis. Many patients had poor adherence and/or multiple changes in prophylactic therapy, complicating the interpretation of our results. These changes and intermittent prophylaxis with other agents likely contribute to both the efficacy and adverse effects observed. Although adherence was suboptimal, these data reflect a real-world population and infusion clinic setting. Encouragingly, despite frequent late doses and missed doses, no breakthrough P. jirovecii infections were identified.

The results of our study support our current practice and indicate that IV pentamidine is an acceptable alternative to TMP-SMX for PJP prophylaxis. Because pentamidine does not provide activity against Toxoplasma, it is important that patients be screened for previous exposure and the results considered before using pentamidine for PJP prophylaxis. For such patients, a change in prophylaxis to TMP-SMX as soon as possible is prudent.

MATERIALS AND METHODS

Eligible patients were identified in the EHR based on utilization of International Classification of Diseases codes for patients ≥18 years of age who received care at Oregon Health and Science University (OHSU) Hospitals and Clinics between January 2007 and September 2017. Inclusion required receiving an allogeneic HSCT at OHSU and at least 1 dose of IV pentamidine post-HSCT for PJP prophylaxis. Data were collected from the first pentamidine dose until PJP prophylaxis was discontinued.

The primary endpoint was the rate of breakthrough P. jirovecii infections in patients actively receiving IV pentamidine as prophylaxis. A diagnosis of PJP was included if it occurred no sooner than 7 days after the initial dose and no later than 28 days after the last dose of pentamidine. A positive result from a sputum or plasma PCR assay, bronchoalveolar lavage fluid staining, or clinical suspicion by the provider was considered diagnostic of P. jirovecii infection. Secondary endpoints included the incidence of adverse effects attributed to pentamidine infusions, which were identified from notes in the medical record. The incidence of breakthrough infections caused by Toxoplasma gondii, Listeria monocytogenes, and Nocardia species during IV pentamidine prophylaxis was also determined. Other data collected included cancer diagnosis, concurrent medications, history of P. jirovecii infection, number of pentamidine doses, and details of pentamidine dosing. Doses administered >35 days after the prior dose were late. Doses >28 days late were considered missed doses.