The usual first treatment for Pneumocystis jirovecii pneumonia (PCP) is an antibiotic called trimethoprim-sulfamethoxazole (TMP-SMX). However, some patients cannot take this medication because of allergies, side effects, or lack of response.

This study asks the question:

When TMP-SMX cannot be used, which alternative treatment for PCP provides the best balance of effectiveness and safety?

Pneumocystis jirovecii pneumonia (PCP) is a serious lung infection that affects people with weakened immune systems (e.g., patients with cancer, organ transplants, autoimmune diseases, or HIV). Without timely treatment, PCP can lead to respiratory failure and death.

TMP-SMX is the standard first-line treatment, but 20-30% of patients cannot receive the treatment or cannot tolerate it due to allergic reactions, kidney problems, low blood counts, drug interactions, or treatment failure. In these situations, doctors use alternative medications such as clindamycin with primaquine, pentamidine, or atovaquone.

Although these alternative treatments are widely used, there is limited modern research directly comparing them. As a result, treatment choices vary between hospitals and physicians.

The main objective of this study is to determine which alternative treatment works best for patients with PCP who cannot receive TMP-SMX. Eligible participants in the PCP alternatives therapy are enrolled and randomized centrally 1:1 in the MUHC Research Electronic Data Capture (REDCap) system. The primary outcome is a Hierarchical composite Win Ratio Outcome at day 30: death; new extracorporeal membrane oxygenation (ECMO), new invasive mechanical ventilation; severe (CTCAE grade 4) adverse drug event; and length of stay in hospital (amongst survivors). Secondary endpoints include individual components of the composite outcome, and tertiary endpoints include quality of life and longer-term outcomes through day 180.

Inclusion Criteria:

• Immunocompromised patients (including but not limited to HIV, solid organ transplant, solid tumors, hematological transplant and malignancies, systemic diseases, chemotherapy, long term corticosteroid use, and immunosuppressive therapies, as well as primary immunodeficiencies) in an emergency department, cliinic, or hospital

  • Age ≥18 years
  • Proven or probable Pneumocystis jirovecii pneumonia
  • Inability to receive trimethoprim-sulfamethoxazole due to contraindication, intolerance, toxicity, or treatment failure
  • Immunocompromised status
  • Ability to provide informed consent (or per local regulations)

While participants may be enrolled in multiple domains of the SPIRIT-PCP Platform over time (if they are eligible and a domain is active), they may only be enrolled to single question once (e.g., they can be part PCP Alternatives and an eventual secondary prophylaxis domain; however, if they have a recurrence, they cannot be included in PCP Alternatives again).

Exclusion Criteria:

• The Platform will exclude: patients where we are unable to obtain informed consent, where patients or their proxy have declined to consent, where the treating team has declined participation, where follow up cannot be reliably obtained (e.g., lack of means of communication, patient non-resident of jurisdiction), where treatment with antibiotics is not in keeping with a patient's advanced care directives, and where death is deemed imminent (<48h) as determined by the treating team and site investigator.

Clinical:

1. Previous severe adverse reaction or hypersensitivity to clindamycin, primaquine, or atovaquone (mild-moderate PCP) or to clindamycin, primaquine, or pentamidine (severe PCP);

2. More than 7 calendar days of any therapy for PCP (no more than 4 can involve a study drug).

3. Known pregnancy or breastfeeding (pregnancy test will be offered)

   Drug specific exclusion criteria:

4. For clindamycin-primaquine:

   1. Known G6PD deficiency OR family history of G6PD deficiency without excluding by testing*
   2. Known diagnosis of porphyria
   3. Concomitant use of methotrexate or cyclophosphamide which cannot be held *G6PD deficiency is an X-linked recessive genetic disease. Female patients without a family history are very unlikely to have this disease and so therapy can start while waiting for the test in the absence of a family history. Male patients should wait for test results prior to receiving primaquine even if they do not have a family history. For those without G6PD testing at diagnosis, it is a reasonable standard of care to order testing.

5. For pentamidine:

   1. Absence of adequate intravenous access as determined by treating team and site investigator. In the event of loss of IV, up to 2 consecutive doses can be given intramuscularly if the patient is not systemically anticoagulated and does not have a coagulopathy.
   2. Hypotension defined as systolic blood pressure below 90mmHg without pharmacologic support
   3. Personal history of Torsade de Pointes or presence of a corrected QTc of greater than 490ms on ECG on date of enrollment

6. For atovaquone:

   1. Receipt of PCP Prophylaxis (≥3 doses per week) for ≥ 4 weeks with atovaquone
   2. inability to tolerate atovaquone with a meal or enteral feeding (e.g., prolonged NPO status is an exclusion as atovaquone must be taken with food for proper absorption)
   3. Concurrent use of rifampin, rifabutin, or tetracycline (that cannot be stopped)
   4. Reduced gastric absorption (patient must not have a medical condition which the treating team and/or site investigator believes will interfere with atovaquone absorption, e.g., total gastrectomy)

Administrative:

1. Trial site not participating in PCP Alternatives branch of the initial therapy domain

Individual anonymized patient data will be shared for one year, released one year following the publication of the primary PCP Alternatives Manuscript. Data will be accessible through reasonable request to Emily McDonald at emily.mcdonald@mcgill.ca and with an inter-institutional agreement in place.

Contact emily.mcdonald@mcgill.ca; scientists seeking to answer secondary questions about the data or combine the data in a systematic review or meta-analysis. They can access anonymous individual patient data. There will need to be an inter-institutional data sharing agreement in place.