Comprehensive review of safety in Experimental Human Pneumococcal Challenge

Abstract

Introduction: Experimental Human Pneumococcal Challenge (EHPC) involves the controlled exposure of adults to a specific antibiotic-sensitive Streptococcus pneumoniae serotype, to induce nasopharyngeal colonisation for the purpose of vaccine research. The aims are to review comprehensively the safety profile of EHPC, explore the association between pneumococcal colonisation and frequency of safety review and describe the medical intervention required to undertake such studies.

Methods: A single-centre review of all EHPC studies performed 2011-2021. All recorded serious adverse events (SAE) in eligible studies are reported. An unblinded meta-analysis of collated anonymised individual patient data from eligible EHPC studies was undertaken to assess the association between experimental pneumococcal colonisation and the frequency of safety events following inoculation.

Results: In 1416 individuals (median age 21, IQR 20-25), 1663 experimental pneumococcal inoculations were performed. No pneumococcal-related SAE have occurred. 214 safety review events were identified with 182 (12.85%) participants presenting with symptoms potentially in keeping with pneumococcal infection, predominantly in pneumococcal colonised individuals (colonised = 96/658, non-colonised = 86/1005, OR 1.81 (95% CI 1.28-2.56, P = <0.001). The majority were mild (pneumococcal group = 72.7% [120/165 reported symptoms], non-pneumococcal = 86.7% [124/143 reported symptoms]). 1.6% (23/1416) required antibiotics for safety.

Discussion: No SAEs were identified directly relating to pneumococcal inoculation. Safety review for symptoms was infrequent but occurred more in experimentally colonised participants. Most symptoms were mild and resolved with conservative management. A small minority required antibiotics, notably those serotype 3 inoculated.

Conclusion: Outpatient human pneumococcal challenge can be conducted safely with appropriate levels of safety monitoring procedures in place.

Fig 1. EHPC study design and safety procedures.

A) A schematic demonstrating EHPC study timepoints and procedures. Interventions including vaccination or further pneumococcal inoculation can occur at pre-specified study visits. Experimental Colonisation status is identified in nasal wash samples taken at minimum of day 2 and day 7 post-inoculation. Participants are monitored for symptoms from inoculation until their final visit. B) Overview of safety procedures in EHPC. A schematic diagram summarising the safety processes in place prior to participant inclusion in the study, then following inoculation and post-study. Regulatory processes are not included in this diagram.

Fig 2. Study size, number of inoculations per study and frequency of safety review for potential pneumococcal symptoms.

The stacked bar graph demonstrates the percentage of participants reporting symptoms in each study (pink-symptoms, green- no symptoms). The bar chart demonstrates the number of inoculations performed in each study. Most EHPC studies involved one inoculation per participant, 4 studies involved two or more inoculations per participant and one study involved three inoculations per participant.

Fig 3. Comparison of reported symptom onset time in colonised and non-colonised individuals.

A) Violin plots demonstrating the onset time for the first reported symptom following inoculation for colonised (orange) and non-colonised individuals (green). Each study is identified in a different colour. B) Stacked bar graph demonstrating the percentage of reported symptoms of the total that have occurred by each timepoint (days post-inoculation) in those colonised and non-colonised.

Fig 4. Forest plots illustrating the OR for safety review post-inoculation for those pneumococcal colonised and non-pneumococcal colonised by study.

A) Forest plot of all safety reviews (combining potentially pneumococcal and non-pneumococcal related) for all included EHPC studies using an unadjusted random effects model. Effect sizes are reported as OR with 95% CI. Studies are ranked by their OR. The overall OR is shown in bold, and a dotted line is included for comparison across studies. B) Forest plot of all potentially pneumococcal safety reviews for all EHPC studies using an unadjusted random effects model, adjusted for inoculum serotype (SPN6B, SPN23F+SPN6B, SPN15B/SPN23F +SPN6B, SPN3).

Fig 5. Odds ratio plot of the most frequently reported symptoms in keeping with pneumococcal disease reported at safety review.

A) OR (95% CI) of each potentially pneumococcal symptom occurring in experimentally colonised individuals compared to non-colonised post-inoculation B) OR (95% CI) of each potential pneumococcal symptom occurring in experimentally colonised individuals compared to non-colonised, excluding SPN3 inoculated participants. Potential pneumococcal symptoms were pre-identified through a literature search (results shown in S3 Table). OR were calculated using Fisher’s Exact Test. A dotted line identifies an OR of 1 for comparison between symptoms. Only ‘sore throat’ and ‘malaise’ are statistically significant, with only ‘malaise’ significant when SPN3 inoculated individuals are removed.

Fig 6. Summary of a common protocol features for safely conducting EHPC in an outpatient setting.

A schematic demonstrating the key features of a common protocol for safely performing EHPC pre- and post-pneumococcal inoculation.