Introduction
Pneumocystis pneumonia (PCP) is caused by
Pneumocystis jirovecii, which was previously classified as a protozoan, but is currently considered to be a fungus on the basis of nucleic acid and biochemical analysis. PCP is one of the commonest opportunistic infections in patients with HIV/AIDS, often occurring in patients with CD4
+ T cell counts less than 200 cells/μL, and is one of the main causes of hospitalization and death among patients with HIV/AIDS. Some early studies showed that the mortality of HIV-infected patients with severe PCP was 36–50% in the early antiretroviral therapy (ART) era [
1]. A past study showed that in patients with HIV/PCP who had received ART, mortality was 9.9%; in those who had not received ART, mortality was 12.0% [
2]. In the modern ART era, the mortality rate in HIV-infected patients with PCP has decreased to approximately 10–12%; however, mortality can rise to up to 84% in those with moderate to severe disease without any treatment [
3]. Therefore, early selection of effective treatment is a priority in these patients.
Trimethoprim/sulfamethoxazole (TMP-SMX), also known as co-trimoxazole/Bactrim®, is currently recommended as first-line treatment of PCP in HIV-infected patients because of its relatively high overall efficacy and the widespread availability of its oral and parenteral formulations [
1]. TMP-SMX can relieve the symptoms of PCP and improve prognosis. However, the efficacy rate of patients on TMP-SMX monotherapy administered to HIV-infected patients with moderate to severe PCP is 70–80%, and some patients show poor sensitivity to TMP-SMX monotherapy [
4]. TMP-SMX is considered to only destroy trophoblasts, and has no effect on the tomont stage of the organism [
5]. In addition, TMP-SMX is slow to act, requiring approximately 5–8 days to achieve curative effect, which suggests that the use of TMP/SMX may be inappropriate for critically ill patients [
6]. Therefore, the quest for more effective novel therapeutic regimens to improve the prognosis of patients with moderate to severe PCP is an important and currently relevant issue to be addressed [
7].
Clindamycin belongs to the lincosamine class of antibiotics, and can act as an alternative to TMP-SMX in terms of therapeutic success and patient safety [
8]. Used together with primaquine, clindamycin has been suggested for patients who have moderate to severe PCP in the current US Department of Health and Human Services (DHHS) guideline [
9]. Because primaquine is not available in China owing to the elimination of malaria there, it has become necessary to seek a new therapeutic option to replace it. Not many reports in the literature exist with regards to the combination of TMP-SMX with clindamycin [
10], even though in clinical practice the efficacy of TMP-SMX plus clindamycin appears to be greater than that of TMP-SMX monotherapy.
In recent years, an increasing number of studies report that caspofungin may be used to treat PCP [
11]. However, it is believed at present that caspofungin alone should not be used for the treatment of PCP (especially in moderate to severe PCP), as it is unlikely to completely clear
Pneumocystis, and may allow PCP recurrence after drug withdrawal [
12]. TMP-SMX exerts its effects on trophic forms of
P. jirovecii, while caspofungin primarily acts on cystic forms of
P. jirovecii [
13], and the combination of caspofungin and TMP-SMX thus has the potential to inhibit the entire life cycle of the
Pneumocystis organism [
14]. Case reports have described the clinical success of caspofungin when used in combination with TMP-SMX in treating moderate to severe PCP in immunocompromised hosts [
15]. A clinical trial involving caspofungin salvage treatment in HIV-infected patients with PCP has shown an appreciably high success rate (80%, 8/10 patients) [
13]. Jin et al. found that HIV-negative patients with moderate to severe
P. jirovecii pneumonia had good response to caspofungin combined with TMP-SMX [
16]. In further studies, four patients with PCP were successfully treated with caspofungin and TMP-SMX [
17]. Caspofungin may thus be a promising drug for use in patients with HIV/PCP; however, clinical experience with its use is currently limited.
In our multicenter, prospective observational cohort study of data from 16 hospitals throughout China, we aimed to observe and compare the clinical effectiveness and safety of TMP-SMX monotherapy, TMP-SMX plus clindamycin, and TMP-SMX plus caspofungin for the treatment of moderate to severe PCP (defined by the presence of an arterial oxygen pressure of less than 70 mmHg or an arterial-alveolar gradient of more than 35 mmHg in a patient with HIV/PCP) in HIV-infected patients.
Methods
Patient Population and Study Design
The present multicenter, observational cohort study was conducted at 16 hospitals in China, from January 2019 to December 2020. Our study received human research ethics approval (Approval No. 2019-003-02-KY) from the Ethics Committee of the Chongqing Public Health Medical Center, and from the individual institutional ethics committees of each of the other 15 hospitals involved in this study (Table S1 in the supplementary material), and was duly registered at the Chinese Clinical Trial Registry (Registration No. ChiCTR1900021195). The study was conducted in accordance with the tenets of the 1964 Declaration of Helsinki. All enrolled patients provided written informed consent.
Inclusion and Exclusion Criteria
Patients with confirmed PCP were included in our study if they satisfied the following eligibility criteria: (1) over 18 years old; (2) HIV positive; (3) willing to sign informed consent; (4) moderate-to-severe PCP.
Patients were excluded from this study if they (1) were intolerant of or had severe allergy to any of the therapeutic drugs used; (2) had hemoglobin (HGB) levels less than 60 g/L, white blood cell counts (WBC) less than 1.0 × 109/L, neutrophil counts (N) less than 0.5 × 109/L, platelet counts (PLT) less than 50 × 109/L, blood amylase (AMS) more than two times the upper normal limit (UNL), serum creatinine (Scr) more than 1.5 times UNL, aspartate aminotransferase (AST)/alanine aminotransferase (ALT)/alkaline phosphatase (ALP) more than five times UNL, total bilirubin (TBIL) more than two times UNL, serum creatine phosphokinase (CK) more than two times UNL; (3) had the presence of other serious disease that could have affected the accurate evaluation of efficacy and prognosis; (4) were pregnant or breast-feeding women; (5) had severe mental health illnesses, or used intravenous recreational drugs, (6) had non-Chinese nationality; (7) had mild PCP; or (7) withheld informed consent.
Definitions
Moderate to severe of PCP was defined as: (1) partial arterial oxygen pressure (PaO2) ≤ 70 mmHg, or while
breathing room air, an alveolar-arterial oxygen difference of (A-aDO2) ≥ 35 mmHg; (2) the presence of
relevant pulmonary symptoms, including dry cough, shortness of breath, progressive dyspnea, and purpura,
and may also have fever; (3) pulmonary infiltrations indicated by chest radiography or computed
tomography (CT) performed when PCP was clinically suspected following hospitalization; (4) microbiological
confirmation by positive PCR and/or Grocott’s methenamine silver (GMS) results for sputum, pulmonary
aspirate, or broncho-alveolar lavage fluid (BALF) samples. If the subject met criteria (1), and (2), and (3),
and/or (4), moderate to severe of PCP was assumed to be present. PaO2 was measured directly via blood
gas analysis of arterial blood.
Overall positive response rate was defined as having fewer clinical symptoms, improved PaO2, and resolution
of evidence of pneumonitis on chest imaging after treatment.
Treatment failure was defined as persistent fever and worsening hypoxia, and/or radiographic deterioration.
Data Collection and Quality Assurance
Based on a standard operating procedure (SOP) manual, investigators were trained to ensure patient
adherence to the study protocol. Data including clinical characteristics, efficacy, safety, and adverse effects
were collected at each follow-up visit during the follow-up period. All raw data were recorded in case report
forms (CRFs) and then transferred to an electronic database via the Medical Research Platform. Withdrawals
from the study, or missed visits were fully explained on CRFs. The study monitor reviewed the completed
CRFs quarterly to ensure accuracy and diligence of the application of inclusion, exclusion, and withdrawal
criteria, as well as to ensure that information on the CRFs were consistent with those in the source electronic
medical records.
Treatment and Outcomes
All eligible patients were divided into three groups: Group 1, receiving TMP-SMX monotherapy, Group 2,
receiving TMP-SMX plus clindamycin, and Group 3, receiving TMP-SMX plus caspofungin. During the study
period, we used similar protocols for drug dosage and administration as recommended by international
guidelines. TMP-SMX was given at a daily dose of 15–20 mg/kg of trimethoprim and 75-100 mg/kg of
sulfamethoxazole. Clindamycin was given by intravenous injection at dose of 0.6 g every day. Patients
received 70 mg caspofungin intravenously on the first day, and were then subsequently administered a dose
of 50 mg/day. All treatment regimens lasted for 3–4 weeks. Adjunctive corticosteroids at the dose
recommended by guidelines were administered to all patients. Patients received additional nasal cannula
oxygen therapy, and non-steroidal anti-inflammatory drugs (NSAIDs) as required, as well as oral or
intravenous rehydration, electrolyte correction, antipyretics, analgesics, and antiemetic drugs as their
individual clinical conditions demanded.
The primary outcome was the difference in overall mortality in the three groups at week 4 and week 12.
Secondary outcomes included the proportion of overall positive response to treatment of moderate to severe
PCP in each group at week 4 and week 12, the differences among the three groups in the duration of
treatment, and the difference in rates of adverse events among the three groups during the study period.
Study Procedures
After HIV/PCP patients were admitted to hospital, we selected those patients whom we assessed may be
eligible for our study, and sought their consent for eligibility screening. After they provided their informed
consent, we assessed their eligibility for study inclusion, and only included into our study those who met our
inclusion criteria but who also did not meet the exclusion criteria of our study. Study visits occurred at week
1, week 2, week 3, week 4, and week 12 after initiation of PCP treatment. Safety was assessed by
interrogation of the participants for potential development of clinical symptoms, and physical examination for
clinical signs, clinical laboratory tests, and documentation of adverse events. Treatment adherence was
assessed by review of clinical diaries that were filled out by medical staff.
Statistics
Continuous variables were described as mean (± SD) for normally distributed data, or median with interquartile
ranges (IQR) for non-normally distributed data. Categorical variables were expressed as frequency
rates and percentages. Kolmogorov–Smirnov tests and ANOVA (analysis of variance) tests were used for
continuous variables. Chi-squared tests, continuity corrections, and Fisher’s exact tests were used to test
statistical significance for categorical data. Cox proportional-hazards model were used to analyze and
compare survival rates. All statistical analyses were performed using Statistical Package for the Social
Sciences (SPSS) software, version 25.0 software (IBM SPSS, Armonk, New York, USA), using a statistical
significance threshold of p < 0.05.
Sample size was calculated by using the PASS (Power Analysis and Sample Size) software, version 15 (NCSS,
LLC, Kaysville, Utah, USA), according to the following assumptions: a 12-week mortality rate of 30%, 15%,
and 10% in the TMP-SMX monotherapy group, the TMP-SMX plus clindamycin group, and the TMP-SMX plus
caspofungin group, respectively, with an overall two-sided alpha level of 0.05, and a statistical power of
90%. Thus, a sample size of 263 patients was required for our study. Considering a 10% withdrawal rate,
we planned to enroll 300 participants, and thus have at least 270 participants for analysis.
Discussion
Currently, the first choice of therapeutic drug for the prevention and treatment of PCP is TMP-SMX. Studies have shown that the survival rates of HIV-infected patients with PCP are only about 70% when using TMP-SMZ alone as first-line therapy, and survival rates using TMP-SMZ are even lower in moderate to severe cases. Thomas et al. reported an overall survival rate of 93%, and a survival rate of 81% in patients with severe disease, among 73 HIV-infected patients treated with TMP-SMX [
18]. Some moderate to severe cases can also prove fatal despite full supportive intensive care with mechanical ventilation. This observation is also supported by the results of our prospective study. In our study, the mortality rate of patients with moderate to severe PCP and HIV was 13.04% in the TMP-SMX monotherapy group. We observed that the TMP-SMX plus clindamycin group and the TMP-SMX plus caspofungin group showed no further decrease in mortality compared with those prescribed TMP-SMX only.
Previous case reports showed that the efficacy of caspofungin was favorable when it was used as a first-line drug or for salvage therapy for PCP [
19]. A trial involving the caspofungin salvage therapy for PCP showed a high success rate (80%) among HIV-infected patients [
20]. In contrast, Kamboj et al. described four HIV-negative patients with PCP that had no observable clinical response to caspofungin as salvage therapy [
21]; however, their study is consistent with ours. We found that both therapeutic combinations (TMP-SMX plus clindamycin and TMP-SMX plus caspofungin) demonstrate similar efficacy for the treatment of moderate to severe PCP as TMP-SMX monotherapy. In recent years, the abuse of antibiotics has increased, which not only facilitates the emergence of antibiotic drug resistance but also results in large amounts of wasted antibiotics and of antibiotics entering the water table and waterways, causes an unnecessary economic burden on patients, and causes unnecessary adverse reactions. Currently, China has come to inherit the unfortunate distinction of becoming one of the world’s worst abusers of antibiotics [
22]. Hence, if TMP-SMX monotherapy can be demonstrated to be as effective and efficient as combinations with other antibiotics and antifungals (as has been demonstrated with the results of our study) for the treatment of moderate to severe PCP, then obviously combinations of TMP/SMX with other antibiotics and antifungals should not be used.
The adverse events of TMP-SMX have been widely reported, and include rashes, fever, gastrointestinal complications, bone marrow suppression, hyperkalemia, hepatoxicity, and renal insufficiency [
23]. The adverse effects of clindamycin and caspofungin are similar to those of TMP-SMX. In our study, we there was a high incidence of bone marrow suppression, hepatic dysfunction, and electrolyte disturbances, both at 4 weeks and at 12 weeks, in all three groups, and there were no significant differences in adverse effect rates among the three groups, suggesting that combination therapy with TMP-SMX and clindamycin or caspofungin did not increase the incidence of adverse events in subjects compared with TMP-SMX monotherapy. In addition, we failed to identify individual cases of interstitial nephritis, pancreatitis, and aseptic meningitis in our cohort.
Our study had a few limitations. Firstly, not all patients with moderate to severe PCP underwent bronchoscopic examination and bronchoalveolar lavage for definitive PCP diagnosis because hypoxia and dyspnea in some patients made it technically difficult and potentially hazardous to undertake those procedures; nevertheless, we adopted a unified clinical diagnostic standard for patients with moderate to severe PCP who could not undergo those procedures. Secondly, a limited sample size and the non-randomized nature of our study may have introduced a degree of bias into our study. Also, selection and unmeasured confounding bias cannot be completely excluded. Novel therapeutic interventions should ideally be assessed via randomized, controlled clinical trials. However, such an approach may not be practically feasible in the context of patients with critical illness, especially in immunocompromised patients. We endeavored to include patients at a 1:1:1 ratio, so as to reduce inherent bias caused by the subjective intentionality of clinicians in the study. Thirdly, as a result of TMP-SMX drug resistance being rarely reported in China, we did not investigate the occurrence of drug resistance in P. jirovecii in this study.
Conclusion
Our results indicate that there are no significant differences among the three different treatment regimens in terms of antifungal effectiveness in HIV-infected patients with moderate to severe PCP. Caspofungin and clindamycin are expensive, may not be readily available to patients, and are generally used to manage quite specific infectious agents and diseases, while TMP/SMX is inexpensive, readily available, and is commonly used. TMP-SMX monotherapy as a therapeutic drug regimen to treat HIV-infected patients with moderate to severe PCP is therefore an economically viable, convenient, and appropriate treatment strategy in resource-limited settings.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.