Safety and efficacy evaluation of low-dose trimethoprim-sulfamethoxazole for prophylaxis of Pneumocystis pneumonia in HIV uninfected patients undergoing hemodialysis: a retrospective observational study

Pneumocystis pneumonia (PCP) is a potentially life-threatening infection that occurs in both patients with human immunodeficiency virus (HIV) infection and those who are immunocompromised without HIV infection (HIV-uninfected) [1]. The mortality rate of PCP in immunocompromised HIV-uninfected patients is especially high, at 35–50% [2‐5]_. Therefore, certain guidelines recommend the use of PCP prophylaxis in immunocompromised HIV-uninfected patients [1, 6, 7]. Trimethoprim-sulfamethoxazole (TMP-SMX) has been considered the first-line prophylactic regimen for PCP according to several guidelines [1, 6, 8, 9].

The standard dosage of TMP-SMX for PCP prophylaxis is TMP-SMX 80 mg/400 mg or TMP-SMX 160 mg/800 mg, either daily or three times a week, i.e., 6–14 single strength (SS) tablets/week is considered the standard dosage [1]. Additionally, trimethoprim and sulfamethoxazole are excreted renally; therefore, patients with impaired renal function exhibit an increase in the half-lives of both components, requiring dosage adjustment of the drugs based on creatinine clearance [1, 9]. Moreover, the recommended prophylactic dose of TMP-SMX has been determined based on studies in patients with normal renal function, and little evidence is available regarding the appropriate prophylactic dosage of TMP-SMX against PCP in immunocompromised HIV-uninfected patients undergoing hemodialysis. Therefore, the aim of this study was to investigate the efficacy and safety of low-dose TMP-SMX in patients undergoing hemodialysis.

This single center, retrospective, observational study included 116 HIV-uninfected adult patients who had undergone hemodialysis at Kobe City Medical Center General Hospital between January 1, 2012 and December 31, 2017, and were administered TMP-SMX for prophylaxis against PCP. Among them, we excluded patients who were under a short observation period of less than 14 days (n = 35). The patients were divided into two groups, standard-dose group and low-dose group, according to the TMP-SMX dose administered. The standard dose was defined as ≥6 SS tablets/week and low-dose was defined as < 6 SS tablets/week [1, 10]. With regards to underlying diseases, leukemia, lymphoma, and multiple myeloma were collectively defined as hematological malignancies. Nephrosis, polyangiitis, connective tissue disease, and idiopathic thrombocytopenic purpura (ITP) were collectively defined as autoimmune diseases. Corticosteroid doses were expressed in Prednisolone (PSL) equivalents. If the doses of corticosteroids were tapered during the observation period, the initial dose was defined as the steroid dose. Further, if steroids were administered only on certain days during a period of time, such as when chemotherapy and corticosteroid were used together, the average daily dose during the period was defined as the corticosteroid dose.

The severity of adverse events (AEs) was determined using the Common Terminology Criteria for Adverse Events (CTCAE), version 5.0 [11]. The study was conducted in accordance with the principles of the Declaration of Helsinki. This study was approved by the ethics committee in Kobe City Medical Center General Hospital, and the reference number was zn190616. Informed Consent was waived by the ethics committee in Kobe City Medical Center General Hospital because this study used retrospective data obtained from hospital records and there were no interventions in the study patients.

The cumulative incidence of PCP and cumulative discontinuation rate of TMP-SMX due to AEs were set as endpoints; death was considered as a competing risk.

All statistical analyses were performed with EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R [12].

A total of 81 patients were included in the study as follows: 36 patients in the low-dose group and 45 patients in the standard-dose group. The median age was 67 years (IQR: 60–76 years), and 52 patients (64.2%) were men. Regarding underlying diseases, 23 patients had hematological malignancies, 45 had autoimmune diseases, and 13 had other diseases. Among the patients with hematological malignancies, five had leukemia, nine had lymphoma, and nine had multiple myeloma. Among those with autoimmune diseases, 10 had connective tissue disease, 17 had polyangiitis, 17 had nephropathy, and one had ITP. Among those with other diseases, six had pulmonary disease, three had eosinophilia, and one patient each had drug eruption, adrenocortical insufficiency, encephaloma, and aldosteronism. Most patients with hematological malignancies received chemotherapy (95.6%); 14 patients among them were treated with a regimen of chemotherapy including corticosteroids. Furthermore, most patients (97.8%) with autoimmune disease received corticosteroids; 30 patients among them received steroids alone while 14 patients received a combination of steroids and immunosuppressive drugs. The median dose of corticosteroid was 30 mg (IQR: 20-40 mg). The baseline characteristics of the patients and the breakdown of immunosuppressive treatments administered to each patient are described in Table 1. There were no significant differences in age, sex, weight distribution, or treatment between the two groups. Particularly, the low-dose group included more patients with hematological malignancies and fewer patients with autoimmune disease than the standard-dose group.

The median observation period of patients without events was 187 days (IQR: 60–458 days) in the low-dose group and 201 (IQR: 49–459 days) in the standard-dose group (P = 0.586). None of the patients in either group developed PCP during the observation period. The cumulative incidence of discontinuation was 12.1% (95% CI: 0.027–0.29) in the low-dose group and 35.6% (95% CI: 0.20–0.52) in the standard group (P = 0.019, Fig. 1). The aHR of the low-dose group compared to that of the standard-dose group was 0.18 (95% CI: 0.04–0.86, P = 0.032) (Table 2).

The most frequent AE that caused TMP-SMX discontinuation in the low-dose group was thrombocytopenia and leukocytopenia, while in the standard-dose group, it was rash, thrombocytopenia, anemia, and hyponatremia (Table 3).

In the present study, we retrospectively compared the efficacy and safety of low-dose and standard-dose TMP-SMX for prophylaxis against PCP in 81 immunocompromised HIV-uninfected patients undergoing hemodialysis. With regards to efficacy, no patients developed PCP in both the low-dose and standard-dose groups. Additionally, the discontinuation rate of TMP-SMX regimen among the low-dose group was lower than that in the standard-dose group, suggesting that a low-dose of TMP-SMX may be better tolerated than standard-dose in patients undergoing hemodialysis.

Several studies have compared the therapeutic and prophylactic effects of low-dose and standard dose of TMP-SMX. Prophylaxis of PCP with 6–14 SS tablets/week of TMP-SMX, was associated with a similar development rate of PCP in several reports. Therefore, the guidelines recommend 6–14 SS tablets/week of TMP-SMX for prophylaxis of PCP. In addition, low-dose TMP-SMX was administered to patients with normal renal function. For example, an open-label randomized control study of patients with rheumatoid arthritis revealed that the development of PCP was similar between the standard-dose group (7 SS/week) and low-dose group (3.5 SS/week) [13]. Another retrospective study of patients who underwent allogeneic hematopoietic stem cell transplantation revealed a similar efficacy between low-dose (4 SS/week) and standard-dose TMP-SMX (7 SS/week) in 156 patients [13, 14]. These studies included a few patients with creatinine clearance of < 30 mL/min; however, no patients were undergoing hemodialysis. The prophylactic efficacy of TMP-SMX was expected to be similar or higher in patients with impaired renal function than in patients with normal renal function because the drugs were renally excreted and half-lives of both components increased in patients with impaired renal function. As expected, our study revealed that the development of PCP was not observed in either the low-dose (5 SS/week) or standard-dose (6–14 SS/week) TMP-SMX groups. These results suggest that low-dose TMP-SMX administration may be an option for patients undergoing hemodialysis.

Serious AEs have often resulted in the discontinuation of TMP-SMX regimen, and several trials have been conducted to explore safer regimens that will allow continuation of treatment or prophylaxis. A Cochrane review of four trials comparing TMP-SMX prophylaxis (6–14 SS/week) versus placebo or no intervention showed no significant difference in the rate of AEs between the two groups, suggesting the safety of prophylactic TMP-SMX in patients with normal renal function [15]. Moreover, some studies investigating a lower dose of TMP-SMX for prophylaxis of PCP reported that the low-dose TMP-SMX (6 SS/week) group had a lower discontinuation rate than the standard-dose TMP-SMX (7–14 SS/week) group. For example, the discontinuation rates attributable to adverse events were significantly lower in the half-strength group (19.1%) than in the single strength group (41.8%) in patients with rheumatoid arthritis [16]. Another study revealed that the discontinuation rate with low-dose TMP-SMX prophylaxis due to adverse events was low (1.3%) in patients after allogeneic hematopoietic stem cell transplantation [13]. In addition, yet another study revealed that the discontinuation rate with a single strength TMP-SMX tablet three times a week for a year, due to adverse events was 10% in kidney transplant recipients [17]. Similar to these studies, our results showed that the cumulative discontinuation rate was lower in the low-dose group (12%) than that in the standard-dose group (36%). Our results suggest that by adjusting the dosage of TMP-SMX to < 6 SS tablets/week, patients undergoing hemodialysis may be able to continue with the regimen as safely as patients with normal renal function. According to the above results, we considered that low-dose TMP-SMX is an appropriate dose that could maintain a balance between the prophylactic effects for PCP and therapeutic safety.

Our study has several limitations. First, this was a single-center, retrospective, observational study and our data were insufficient to eliminate confounding bias. Second, the incidence of PCP is rare in this population; therefore, the results of this study might lack statistical power for detecting the development of PCP. However, in terms of safety, we found that most of the treatment interruptions due to adverse events occurred early, within the first 12 months after initiation. Hence, the results of this study suggest that a lower dose regimen is expected to be safer compared to a higher dose regimen. Third, we could not perform drug concentration monitoring post hemodialysis, would have more accurately reflected the effects of hemodialysis. We believe that prospective studies that incorporate drug monitoring are needed in the future. However, data on prophylactic doses of TMP-SMX in patients undergoing hemodialysis are limited; therefore, we believe this study is meaningful in demonstrating that low-dose regimens may be safely continued.

PCP did not develop in either the low-dose or standard-dose TMP-SMX groups, and the cumulative discontinuation rate due to adverse events was significantly lower in the low-dose group compared to that in the standard-dose group. We believe that our findings can assist medical and health care professionals in determining TMP-SMX dosage when considering PCP prophylaxis for patients undergoing hemodialysis.