Safety and Pharmacokinetics of Recombinant Human Plasma Gelsolin in Patients Hospitalized for Nonsevere Community-Acquired Pneumonia

There remains an unmet need to address the substantial morbidity and mortality associated with severe community-acquired pneumonia (sCAP). Recombinant human plasma gelsolin (rhu-pGSN) improves disease outcomes in diverse animal models of infectious and noninfectious inflammation. This blinded dose-escalation safety study involved non-intensive care unit (ICU) patients admitted for mild CAP and randomized 3:1 to receive adjunctive rhu-pGSN or placebo intravenously. Thirty-three subjects were treated: 8 in the single-dose phase and 25 in the multidose phase.

ABSTRACT

There remains an unmet need to address the substantial morbidity and mortality associated with severe community-acquired pneumonia (sCAP). Recombinant human plasma gelsolin (rhu-pGSN) improves disease outcomes in diverse animal models of infectious and noninfectious inflammation. This blinded dose-escalation safety study involved non-intensive care unit (ICU) patients admitted for mild CAP and randomized 3:1 to receive adjunctive rhu-pGSN or placebo intravenously. Thirty-three subjects were treated: 8 in the single-dose phase and 25 in the multidose phase. For the single-dose phase, rhu-pGSN at 6 mg/kg of body weight was administered once. For the multidose phase, a daily rhu-pGSN dose of 6, 12, or 24 mg/kg was given on 3 consecutive days. Adverse events (AEs) were generally mild in both treatment groups irrespective of dose. The only serious AE (SAE) in the single-dose phase was a non-drug-related pneumonia in a rhu-pGSN recipient who died after institution of comfort care. One single-dose placebo recipient had a drug-related AE (maculo-papular rash). In the multidose phase, there were 2 SAEs in 1 placebo recipient, including a fatal pulmonary embolism. In the 18 rhu-pGSN recipients in the multidose phase, there were no serious or drug-related AEs, and nausea and increased blood pressure were each reported in 2 patients. The median rhu-pGSN half-life exceeded 17 h with all dosing regimens, and supraphysiologic levels were maintained throughout the 24-h dosing interval in the 2 highest dosing arms. Rhu-pGSN was well tolerated overall in CAP patients admitted to non-ICU beds, justifying a larger proof-of-concept trial in an ICU population admitted with sCAP. (This study has been registered at ClinicalTrials.gov under identifier NCT03466073.)

INTRODUCTION

Community-acquired pneumonia (CAP) is a common and sometimes life-threatening infection (1–4). CAP is usually caused by bacteria and viruses alone or in combination, although an etiology is often not determined despite extensive evaluation (1). Once the most identified pathogen, Streptococcus pneumoniae has become less prevalent in the conjugate vaccine era (5–7). Although CAP is often successfully treated with antibiotics, nearly 20% of hospitalized CAP patients require intensive care (1, 3). The substantial morbidity and mortality associated with severe CAP (sCAP) have not been adequately addressed. Patients with sCAP are at risk for complications both from ineffectively managed infection and the overzealous inflammatory responses that underlie acute lung injury and sepsis.

Plasma gelsolin (pGSN), an abundant protein in the blood of healthy individuals, is functionally distinct from its cytoplasmic isoform (8–13). Upon injury, pGSN leaves the circulation to scavenge intracellular debris leaked from damaged tissues that interfere with host defenses. Concurrently, pGSN enhances bacterial uptake and killing by macrophages (14–26). Once leaked cytoskeletal actin and nuclear DNA are cleared, free pGSN can bind proinflammatory lipid and peptide mediators to foster resolution of local inflammatory injury and prevent its spread to distant uninvolved organs (12, 22–24, 27). Consequently, pGSN boosts the early innate immune response to clear pathogens at the infected site while tempering the injurious consequences of excessively prolonged or distant inflammation (26, 28–31).

An overexuberant inflammatory response can occur in a variety of respiratory diseases, like influenza or the recently identified coronavirus disease 19 (COVID-19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). COVID-19 patients may suffer from injurious inflammation and/or cytokine storm, and they can die from acute respiratory distress syndrome or endure debilitating pulmonary fibrosis (32–34). pGSN offers the promise of aborting lung injury in these patients.

Circulating pGSN is consumed in serious infectious and noninfectious conditions, such as bacterial sepsis, major trauma, burns, prolonged hyperoxia, and malaria resulting in extensive tissue injury (35–45). Unlike a positive acute-phase reactant, inflammation does not trigger a compensatory upregulation of pGSN. Correlative studies of patients following a diverse spectrum of common insults have established a consistent relationship among the severity of the precipitating insult, the magnitude of the resultant pGSN decline, and the subsequent likelihood of mortality or devastating complications. In particular, patients admitted with CAP who have the lowest pGSN levels at presentation have the worst outcomes (35). Although the degree of depletion of pGSN at hospital admission correlated with worse outcomes on a population level, the distribution of pGSN levels overlaps among prognostic groups, precluding the use of circulating pGSN concentrations for treatment decisions in individual patients (35).

Our ultimate goal is to investigate if supplementing depleted levels of pGSN in cases of sCAP would be beneficial. As an initial step toward this goal, we conducted a dose-finding trial to evaluate the safety and pharmacokinetics (PK) of adjunctive recombinant human plasma gelsolin (rhu-pGSN) in modestly ill patients hospitalized for acute CAP.

(This work was presented at ASM Microbe 2020.)

RESULTS

Participant accounting.

All subjects had both clinical and radiological confirmation of CAP at screening and received standard-of-care treatment including antibiotics in addition to study drug. The correct study drug was given to every randomized participant.

As specified in his belatedly discovered advanced directive, 1 subject was withdrawn from the study in the single-dose phase shortly after receiving the assigned single dose of rhu-pGSN due to worsening CAP and died the following day (Table 1). A placebo recipient in the multidose phase died on day 3 from a suspected pulmonary embolism after receiving 2 of the 3 planned doses. Since this subject did not complete the full course of study treatment, a replacement subject was enrolled into the same placebo group. Other than this subject, every participant received all the assigned doses. Two rhu-pGSN recipients in the single-dose phase declined participation in the PK substudy.

TABLE 1.

Subject disposition for ITT population^a

	No. (%) of subjects
Subject disposition			Multidose
	Single dose		Rhu-pGSN			Combined rhu-pGSN (N = 18)	Placebo (N = 7)
	Rhu-pGSN, 6 mg/kg (N = 6)	Placebo (N = 2)	6 mg/kg (N = 6)	12 mg/kg (N = 6)	24 mg/kg (N = 6)
Randomized	6 (100.0)	2 (100.0)	6 (100.0)	6 (100.0)	6 (100.0)	18 (100.0)	7 (100.0)
Completed the study	5 (83.3)	2 (100.0)	6 (100.0)	6 (100.0)	6 (100.0)	18 (100.0)	6 (85.7)
Discontinued	1 (16.7)	0	0	0	0	0	1 (14.3)
Primary reason for discontinuations
Adverse event	1 (16.7)	0	0	0	0	0	0
Death	0b	0	0	0	0	0	1c (14.3)

^aITT, intention to treat; N, number of subjects enrolled in each treatment group.

^bSubject was withdrawn from the study shortly after discovery of an advance directive on day 1 prescribing no aggressive measures and died the next day.

^cSubject died after only 2 doses of study drug and was replaced per protocol.

Baseline characteristics.

The age of participants in the study ranged between 22 and 83 years (Table 2). The mean age of the participants was comparable between treatment groups in the single-dose phase (62 years for rhu-pGSN versus 64 years for placebo), while it differed in the multidose phase (55 years for rhu-pGSN versus 64 years for placebo). All participants were Caucasian, 6 being recruited in Australia to the single-dose arm and the remaining 27 from the Republic of Georgia. The gender distribution differed between treatment groups overall (58.3% female rhu-pGSN recipients versus 33.3% female placebo recipients).

TABLE 2.

Baseline characteristics for ITT/safety population

Parameter	Single dose		Multidose
			Rhu-pGSN			Combined rhu-pGSN (N = 18)	Placebo (N = 7)
	Rhu-pGSN, 6 mg/kg (N = 6)	Placebo (N = 2)	6 mg/kg (N = 6)	12 mg/kg (N = 6)	24 mg/kg (N = 6)
Age (yr)
Mean (SD)	62.2 (16.6)	64.0 (2.8)	69.3 (11.4)	55.3 (8.8)	40.0 (19.6)	54.9 (18.0)	64.0 (17.6)
Median	63.0	64.0	69.5	53.5	34.0	56.5	77.0
Range	35.0–81.0	62.0–66.0	56.0–83.0	45.0–70.0	22.0–66.0	22.0–83.0	39.0–78.0
Sex, n (%)
Male	4 (66.7)	2 (100.0)	1 (16.7)	3 (50.0)	2 (33.3)	6 (33.3)	4 (57.1)
Female	2 (33.3)	0	5 (83.3)	3 (50.0)	4 (66.7)	12 (66.7)	3 (42.9)
Race, n (%)
White	6 (100.0)	2 (100.0)	6 (100.0)	6 (100.0)	6 (100.0)	18 (100.0)	7 (100.0)
Height (cm)
Mean (SD)	174.33 (7.31)	170.00 (7.07)	165.83 (10.15)	169.83 (8.93)	167.50 (10.48)	167.72 (9.43)	169.71 (6.60)
Median	174.50	170.00	162.50	169.50	165.50	165.00	170.00
Range	165.0–185.0	165.0–175.0	158.0–185.0	157.0–180.0	153.0–180.0	153.0–185.0	162.0–181.0
Weight (kg)
Mean (SD)	85.15 (13.81)	77.80 (11.31)	75.55 (14.35)	79.35 (19.35)	82.62 (13.49)	79.17 (15.27)	73.20 (14.62)
Median	88.50	77.80	80.50	80.40	82.05	82.05	74.20
Range	60.0–96.6	69.8–85.8	55.3–90.0	52.9–99.0	62.2–99.1	52.9–99.1	58.0–99.5
CURB-65 score
Median	1.0	1.0	1.5	0.0	0.0	0.0	1.0
Range	0–2	1–1	0–2	0–1	0–1	0–2	0–2
PSI class, n (%)
I	0 (0.0)	0 (0.0)	0 (0.0)	1 (16.7)	2 (33.3)	3 (16.7)	2 (28.6)
II	3 (50.0)	1 (50.0)	2 (33.3)	3 (50.0)	4 (66.7)	9 (50.0)	1 (14.3)
III	1 (16.7)	1 (50.0)	3 (50.0)	2 (33.3)	0 (0.0)	5 (27.8)	3 (42.9)
IV	2 (33.3)	0 (0.0)	1 (16.7)	0 (0.0)	0 (0.0)	1 (5.6)	1 (14.3)
V	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0	0
SOFA score
Median	1.5	2.0	1.0	0.5	1.0	1.0	1.0
Range	0–4	2–2	0–2	0–1	0–3	0–3	0–4
hsCRP (mg/liter)
Median	326.7 (N = 5)	395.9	62.1	4.8	9.05	11.75	62.3
Range	4.3–554.3	347–444.8	11.6–316.0	0.8–34.4	1.2–192.0	0.8–316.0	0.7–285.7

In the single-dose phase, 6/8 subjects had sputum cultures. Four samples grew upper respiratory tract flora (including 1 subject with a urinary antigen positive for Streptococcus pneumoniae), 1 sample grew Klebsiella species not further identified, and 1 sample grew Streptococcus not otherwise identified. Six subjects had blood cultures, of which one set grew Staphylococcus not further identified. In the multidose phase, 7/25 subjects had sputum cultures, of which 2 grew Candida albicans, 2 grew Streptococcus not further identified, 1 grew Klebsiella not further identified, 1 grew Enterobacter not further identified, and 1 had both Haemophilus parainfluenzae and Neisseria species. Four subjects had blood cultures, all of which were negative.

The median (range) endogenous pGSN levels prior to treatment were 50 (20.6 to 75.1) μg/ml for 9 placebo recipients (2 in the single-dose arm and 7 in the multidose arms), 49.3 (28.4 to 528.3) μg/ml for 10 rhu-pGSN 6-mg/kg recipients (4 in the single-dose arm and 6 in the multidose arm), 63.2 (49.0 to 70.0) μg/ml for 6 rhu-pGSN 12-mg/kg recipients, and 72.6 (28.8 to 100.1) μg/ml for 6 rhu-pGSN 24-mg/kg recipients (median endogenous pGSN levels in 20 healthy individuals were found to be 56.8 μg/ml [interquartile range, 52.6 to 65.4 μg/ml] [35]). The high maximum in the rhu-pGSN recipients was due to one outlier in the single-dose cohort whose baseline concentration was >10 times higher than the median value (49.3 μg/ml) of the 6-mg/kg group.

Subjects generally had low disease severity scores at entry (Table 2). Procalcitonin levels were below the detection limit in 69% of the tested individuals on day 1, while 85% of all measurements remained below detection limits at later time points. Median baseline levels of high-sensitivity C-reactive protein (hsCRP) ranged from 4.8 to 395.9 mg/liter across arms. The illness severity threshold for hospital admission in Australia may be higher than that in the Republic of Georgia. Six of the 8 patients in the single-dose arm were recruited in Australia, while the remaining patients (including all the subjects in the multidose arms) were recruited from Georgia. A practice difference between countries could explain why the hsCRP levels at admission were higher in the single-dose than multidose cohorts (Table 2). Changes from baseline in severity scores and biomarkers were generally small over the hospitalization.

Treatment-emergent AEs.

In the single-dose phase, 4 of 6 rhu-pGSN recipients (66.7%) experienced 7 adverse events (AEs), and the 2 placebo recipients (100%) experienced 8 AEs (Table 3). Most AEs were grade 1 or 2, but 1 subject in each group reported at least grade 3 AEs. One placebo recipient had a grade 3 tachypnea that was not serious and probably unrelated to study intervention. The only serious AE was reported in a rhu-pGSN recipient who had grade 5 pneumonia, considered unrelated to study intervention but leading to withdrawal and a fatal outcome. No other subjects in the single-dose arm discontinued the study. The sole drug-related AE was a maculo-papular rash in a placebo recipient.

TABLE 3.

Adverse events (AEs)^a for safety population

Parameter	Single dose		Multidose
			Rhu-pGSN			Combined rhu-pGSN (N = 18)	Placebo (N = 7)
	Rhu-pGSN 6 mg/kg (N = 6)	Placebo (N = 2)	6 mg/kg (N = 6)	12 mg/kg (N = 6)	24 mg/kg (N = 6)
No. of AEs	7	8	8	4	3	15	7
No. (%) of subjects with at least 1:
AE	4 (66.7)	2 (100.0)	4 (66.7)	3 (50.0)	2 (33.3)	9 (50.0)	4 (57.1)
NCI-CTCAE grade ≥3 AE	1 (16.7)	1 (50.0)	1 (16.7)	0	0	1 (5.6)	1 (14.3)
Serious AE	1 (16.7)	0	0	0	0	0	1 (14.3)
Drug-related AE	0	1 (50.0)	0	0	0	0	0
AE leading to death	1 (16.7)	0	0	0	0	0	1 (14.3)
Procedure-related AE	0	0	0	0	1 (16.7)	1 (5.6)	0
AE leading to discontinuation	0	0	0	0	0	0	1 (14.3)
No. (%) of subjects with AEs by NCI-CTCAE grade
1	2 (33.3)	1 (50.0)	4 (66.7)	3 (50.0)	2 (33.3)	9 (50.0)	2 (28.6)
2	2 (33.3)	2 (100.0)	1 (16.7)	0	1 (16.7)	2 (11.1)	2 (28.6)
3	0	1 (50.0)	1 (16.7)	0	0	1 (5.6)	1 (14.3)
4	0	0	0	0	0	0	1 (14.3)
5	1 (16.7)	0	0	0	0	0	1 (14.3)
No. (%) of subjects with AEs by relationship to study treatment
Definitely not related	3 (50.0)	2 (100.0)	3 (50.0)	3 (50.0)	0	6 (33.3)	4 (57.1)
Probably not related	2 (33.3)	2 (100.0)	1 (16.7)	0	2 (33.3)	3 (16.7)	0
Possibly related	0	1 (50.0)	0	0	0	0	0
Probably related	0	0	0	0	0	0	0
Definitely related	0	0	0	0	0	0	0
No. (%) of subjects with AEs by relationship to study procedure
Definitely not related	4 (66.7)	2 (100.0)	3 (50.0)	3 (50.0)	1 (16.7)	7 (38.9)	4 (57.1)
Probably not related	1 (16.7)	0	1 (16.7)	0	0	1 (5.6)	0
Possibly related	0	0	0	0	1 (16.7)	1 (5.6)	0
Probably related	0	0	0	0	0	0	0
Definitely related	0	0	0	0	0	0	0

^aGraded according to reference 46. Grade 1, mild; grade 2, moderate; grade 3, severe but not life threatening; grade 4, life threatening; grade 5, death related to AE.

In the combined 3 multidose arms, 9 of 18 rhu-pGSN recipients (50.0%) experienced a total of 15 AEs, and 4 of 7 placebo recipients (57.1%) had a total of 7 AEs (Table 3). Most of the AEs were grade 1 or grade 2. One subject in each group reported at least 1 AE of grade ≥3. A subject who received rhu-pGSN at 6 mg/kg experienced grade 3 pyrexia and grade 3 insomnia. One placebo recipient had grade 3 hypotension, grade 4 pneumonia, and a grade 5 pulmonary embolism, which was the presumed cause of death after receiving 2 of the 3 planned doses. None of the AEs were considered drug-related, although 1 24-mg/kg rhu-pGSN recipient in the multidose phase had a procedure-related tachycardia.

A relationship between dose of rhu-pGSN and number or types of AEs was not evident. The proportion of rhu-pGSN recipients with AEs was highest in the lowest-dose group and lowest in the highest-dose group (66.7% in the 6-mg/kg group, 50.0% in the 12-mg/kg group, and 33.3% in the 24-mg/kg group).

In the single-dose phase, no specific AE was reported in more than 1 subject in either treatment group (Table 4). In the multidose phase, no specific AEs occurred in more than 2 subjects in any treatment group, and those that occurred in 2 rhu-pGSN recipients in this phase were nausea and increased blood pressure.

TABLE 4.

Specific adverse events (AEs)^a for safety population

	No. (%) of subjects
Parameter	Single dose		Multidose
			Rhu-pGSN			Combined rhu-pGSN (N = 18)	Placebo (N = 7)
	Rhu-pGSN 6 mg/kg (N = 6)	Placebo (N = 2)	6 mg/kg (N = 6)	12 mg/kg (N = 6)	24 mg/kg (N = 6)
Subjects with ≥1 AE	4 (66.7)	2 (100.0)	4 (66.7)	3 (50.0)	2 (33.3)	9 (50.0)	4 (57.1)
Leukocytosis	0	0	0	0	1 (16.7)	1 (5.6)	0
Tachycardia	0	0	0	0	1 (16.7)	1 (5.6)	0
Constipation	1 (16.7)	0	0	0	0	0	0
Diarrhea	0	1 (50.0)	0	0	0	0	0
Nausea	0	0	1 (16.7)	1 (16.7)	0	2 (11.1)	0
Fatigue	0	0	1 (16.7)	0	0	1 (5.6)	1 (14.3)
Pyrexia	0	0	1 (16.7)	0	0	1 (5.6)	1 (14.3)
Cellulitis	0	1 (50.0)	0	0	0	0	0
Pneumonia	1 (16.7)	0	0	0	0	0	1 (14.3)
Blood alkaline phosphatase increased	0	0	0	1 (16.7)	0	1 (5.6)	0
Blood pressure increased	0	0	2 (33.3)	0	0	2 (11.1)	0
Blood creatine phosphokinase increased	1 (16.7)	0	0	0	0	0	0
Liver function test abnormal	0	1 (50.0)	0	0	0	0	0
Hyperglycemia	0	0	0	1 (16.7)	0	1 (5.6)	0
Gout	0	1 (50.0)	0	0	0	0	0
Musculoskeletal pain	1 (16.7)	0	0	0	0	0	0
Neck pain	1 (16.7)	0	0	0	0	0	0
Headache	1 (16.7)	0	1 (16.7)	0	0	1 (5.6)	1 (14.3)
Dizziness	0	0	1 (16.7)	0	0	1 (5.6)	0
Neuralgia	0	1 (50.0)	0	0	0	0	0
Insomnia	0	0	1 (16.7)	0	0	1 (5.6)	0
Tachypnea	0	1 (50.0)	0	0	0	0	0
Pulmonary embolism	0	0	0	0	0	0	1 (14.3)
Night sweats	0	1 (50.0)	0	0	0	0	0
Rash maculo-papular	0	1 (50.0)	0	0	0	0	0
Hypertension	0	0	0	1 (16.7)	0	1 (5.6)	1 (14.3)
Hypotension	1 (16.7)	1	0	0	0	0	1 (14.3)

^aDefined according to reference 46.

There were 2 deaths during the study, one each in the rhu-pGSN and placebo groups. Neither was thought to be drug related. A 69-year-old male presenting with progressive CAP associated with pleuritic pain, sepsis, tachycardia, and wheezing was randomized to receive rhu-pGSN at 6 mg/kg in the single-dose phase. Shortly after study treatment was administered, the subject was withdrawn because he had an advanced directive not to receive aggressive care. His condition deteriorated rapidly, and he died the following day due to pneumonia. In the opinion of the investigator, his worsening CAP was definitely not related to study treatment. The other death involved a 46-year-old male with CAP randomized to receive placebo in the multidose phase. After the first dose, he was transferred to the intensive care unit (ICU) for respiratory failure and died on the second hospital day after his second dose due to suspected pulmonary embolism.

There were no clear differences between the treatments over time in mean or median laboratory variables and no consistent trends in subjects with shifts from baseline. Abnormalities in blood counts occurred frequently in both treatment arms but were considered related to the underlying condition or to concomitant medication. Three subjects had clinically significant abnormalities in clinical chemistry variables that were reported as AEs. One placebo recipient in the single-dose phase had clinically significant abnormalities in alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase, albumin, CRP, calcium, creatinine, and liver function tests. One 6-mg/kg rhu-pGSN recipient in the single-dose phase had elevated creatine phosphokinase on day 2. A 12-mg/kg rhu-pGSN recipient in the multidose phase had increased blood alkaline phosphatase on day 4.

No abnormalities in electrocardiograms (EKG) were considered clinically significant by the investigator. There were 3 subjects with a mean QT-corrected Fridericia formula (QTcF) of >450 ms. One 6-mg/kg rhu-pGSN recipient in the single-dose phase had a mean QTcF of 459.7 ms on day 28, which was an increase from the baseline of 11.3 ms and was not considered clinically significant. Another 6-mg/kg rhu-pGSN recipient in the single-dose phase had a mean QTcF of 533.0 ms at baseline. No postbaseline EKG was recorded. One 6-mg/kg rhu-pGSN recipient in the multidose phase (3 doses) had a mean QTcF of 453.7 ms at baseline; the mean QTcF was lower (426.3 ms) at day 28.

Pharmacokinetics.

The secondary objective of this study was to determine the pharmacokinetic profile of rhu-pGSN. The pGSN concentrations were measured using an enzyme-linked immunosorbent assay (ELISA) that does not distinguish endogenous pGSN from exogenous rhu-pGSN, so the values presented are the sum of native and added pGSN. The C_max (maximum observed plasma concentration) and exposure parameters (area under the concentration-time curve from zero to 24 h [AUC_0–t] and AUC_0–8) increased with higher doses (Table 5 shows geometric means). Median T_max (time to maximum observed plasma concentration) was 0.1 h after rhu-pGSN injections, except for 24-mg/kg rhu-pGSN on day 3 (1.1 h). Adjusting for predose gelsolin levels in rhu-pGSN recipients and comparison to levels in the placebo groups over time did not change our conclusion.

TABLE 5.

Pharmacokinetic parameters of pGSN (recombinant plus endogenous) following administration of 6, 12, and 24 mg/kg rhu-pGSN on 3 consecutive dosing days (PK population)^c

Parameter	Value by dose of rhu-pGSN
	6 mg/kg, N = 10 or 6a	12 mg/kg, N = 6	24 mg/kg, N = 6	Placebo, N = 9 or 7b
Day 1
AUC0–t (μg·h/ml)	3,125.7 (25.7)	4,845.9 (15.2)	8,454.6 (16.0)	1,083.8 (36.9)
AUC0–8 (μg·h/ml)	1,337.5 (15.9)	2,188.3 (10.6)	3,638.5 (19.2)	358.3 (39.5)
Cmax (μg/ml)	282.1 (69.9)	351.3 (12.3)	592.5 (21.9)	51.4 (38.1)
Tmax (h)	0.1 (0.0–8.1)	0.1 (0.1–2.1)	0.1 (0.0–24.0)	7.6 (0.0–24.0)
t1/2 (h)	24.7 (10.4–34.5) (N = 6)	18.7 (15.4–44.4) (N = 5)	17.1 (12.6–25.0) (N = 4)	NA
Day 2
AUC0–t (μg·h/ml)	3,224.9 (33.4)	5,704.1 (19.7)	7,230.8 (45.0)	1,324.5 (24.2)
AUC0–8 (μg·h/ml)	1,354.6 (32.7)	2,517.3 (18.5)	3,241.1 (50.0)	466.6 (21.3)
Cmax (μg/ml)	273.6 (43.1)	419.0 (21.9)	542.1 (59.2)	64.8 (21.7)
Tmax (h)	0.1 (0.1–2.0)	0.1 (0.1–0.2)	0.1 (0.0–2.0)	16.0 (0.0–23.8)
t1/2 (h)	30.7 (27.8–33.5) (N = 2)	21.4 (10.4–199.0) (N = 6)	23.8 (16.9–205.1) (N = 5)	NA
Day 3
AUC0–t (μg·h/ml)	3,205.4 (20.3)	6,549.1 (21.3)	8,389.2 (66.9)	1,303.2 (30.5) (N = 6)
AUC0–8 (μg·h/ml)	1,314.0 (29.0)	2,995.1 (20.6)	3,772.1 (84.1)	427.9 (35.5) (N = 6)
Cmax (μg/ml)	232.4 (50.1)	466.7 (17.7)	607.0 (88.9)	61.3 (28.3) (N = 6)
Tmax (h)	0.1 (0.1–24.0)	0.1 (0.1–2.0)	1.1 (0.1–16.0)	4.1 (0.0–16.1) (N = 6)
t1/2 (h)	21.8 (21.8–21.8) (N = 1)	20.0 (15.1–32.8) (N = 4)	19.6 (18.7–27.3) (N = 3)	NA

^aIncludes both single dose (N = 4) and multidose (N = 6) on day 1; N = 6 on day 2 and day 3, unless otherwise indicated.

^bIncludes both single dose (N = 2) and multidose (N = 7) on day 1; N = 7 on day 2 and day 3, unless otherwise indicated.

^cAUC_0–t, area under the plasma concentration-time curve from time zero to 24 h; AUC_0–8, area under the plasma concentration-time curve from time zero to 8 h; C_max, maximum observed plasma concentration; NA, not applicable; t_1/2, elimination half-life; T_max, time to maximum observed plasma concentration. AUC and C_max values are given as geometric means (percent geometric coefficient of variation), and T_max and t_1/2 are given as medians (minimum–maximum).

Median t_1/2 (elimination half-life) for each dose exceeded 17 h, albeit with high variability on day 2 of dosing in the 12- and 24-mg/kg recipients. Accumulation of pGSN was minimal based on the AUC and C_max values on day 3 versus day 1 (ratios were slightly greater than 1).

Outcomes.

Differences in clinical efficacy between treatment groups were not evident. The median duration of hospital stay was comparable between rhu-pGSN and placebo recipients in both the single-dose and multidose phases. In the single-dose phase, the median hospital stay was 91.63 h for rhu-pGSN and 86.68 h for placebo recipients. The subject that died in the rhu-pGSN arm of the single-dose phase spent 40.3 h premortem in the hospital. In the multidose phase, the median length of stay was 105.13 h for the combined rhu-pGSN and 103.75 h for the combined placebo recipients. While none of the subjects in the single-dose phase was admitted to the ICU or was intubated and required vasopressors, the subject that died in the placebo arm of the multidose phase was transferred to the ICU and intubated, and he required vasopressors before he died on day 3 (74 premortem hours of hospitalization).

Most biomarkers were below the limit of detection upon presentation and stayed undetectable for the length of the study. In a few cases, elevated baseline hsCRP and interleukin-6 (IL-6) levels may have returned modestly faster and more stably toward normal in rhu-pGSN relative to placebo recipients (see Fig. S1 in the supplemental material), but these selected observations were at most hypothesis generating.

In the single-dose phase of the study, 2 subjects who received rhu-pGSN and 2 subjects who received placebo had a serum sample taken on day 28 to be tested for anti-drug antibodies (ADAs) against pGSN. These 4 test results were screen negative. In the multidose phase of the study, 11 of the 24 tested subjects had a positive result for ADAs at day 28 by the screening assay (9/18 and 2/6 for rhu-pGSN and placebo recipients, respectively) and underwent confirmatory testing. In total, 3 of the 9 screening positives who had received multiple doses of rhu-pGSN were confirmed positive. Both of the screening positives who had received placebo tested negative in the confirmatory assay. Thus, there were a total of 3 confirmed positive cases: one in the 6-mg/kg cohort and two in the 24-mg/kg cohort. Due to concerns about specificity, 11 predose samples were subsequently tested for ADAs, and 2 were positive only to revert to negative by the end of study (Table 6).

TABLE 6.

Frequencies of anti-drug antibodies (ADAs) at day 28 across conditions

Treatment arm	No.
	Tested	Screened positive	Confirmed positive (%)
Single dose
Rhu-pGSN 6 mg/kg	2a	0	0
Placebo	2	0	0
Multidose
Rhu-pGSN 6 mg/kg	6	1	1 (16.7)
Placebo	2	0	0
Multidose
Rhu-pGSN 12 mg/kg	6	2	0
Placebo	2	2	0
Multidose
Rhu-pGSN 24 mg/kg	6	6	2 (33.3)
Placebo	2	0	0
All multidose
Rhu-pGSN	18	9	3 (16.7)
Placebo	6	2	0
All subjects
Rhu-pGSN	20	9	3b (15)
Placebo	8	2	0

^aResults from 4 subjects given rhu-pGSN in the single-dose cohort are missing.

^bTwo rhu-pGSN recipients were confirmed positive at baseline (predose), but neither was positive at day 28.

DISCUSSION

We explored the safety, tolerability, and pharmacokinetics of rhu-pGSN treatment added to the standard of care in this small, double-blinded, placebo-controlled, dose-escalation study (ClinicalTrials registration no. NCT03466073). The only other trial of intravenous (i.v.) rhu-pGSN (Critical Biologics Corporation) was a randomized, double-blind, placebo-controlled, dose escalation study of 28 ICU patients with decreased endogenous gelsolin levels at entry conducted in Hong Kong. There were no restrictions as to the underlying comorbidities. All patients received rhu-pGSN or placebo on top of standard-of-care measures. The four cohorts sequentially received different doses of rhu-pGSN via i.v. infusion, ranging from a single infusion of 3 mg/kg rhu-pGSN up to daily infusions of 6 mg/kg rhu-pGSN for 3 days. During the study, 7/21 rhu-pGSN recipients (33%) died versus 0/7 placebo recipients (0%). None of the deaths were considered related to study treatment. The rhu-pGSN recipients were overall more severely ill than the placebo patients. Transient improvement was observed in a few patients shortly after their rhu-pGSN infusion.

To establish the safety profile of rhu-pGSN in a population less confounded by comorbidities, we conducted our study in mildly ill patients hospitalized with CAP outside an ICU. A total of 33 subjects were treated, including a replacement of one placebo recipient who died in the multidose phase. Overall, preinjection pGSN concentrations were only modestly depleted, reflecting the mild illness in most subjects. Rhu-pGSN was generally safe and well tolerated. There was no correlation between the type and frequency of AEs and the dose of rhu-pGSN. No specific AE was reported in more than 1 subject in the highest-dose group. No specific AEs were reported in more than 2 subjects. The proportion of rhu-pGSN recipients with AEs was highest in the lowest-dose group (4 subjects, 66.7% rhu-pGSN at 6 mg/kg) compared with the middle-dose group (3 subjects, 50.0% rhu-pGSN at 12 mg/kg) and the highest-dose group (2 subjects, 33.3% rhu-pGSN at 24 mg/kg). One subject who received placebo in the single-dose phase had a drug-related maculo-papular rash. None of the AEs in the rhu-pGSN arms were considered related to drug treatment. Two subjects died during the study: one rhu-pGSN recipient in the single-dose phase and one placebo recipient in the multidose phase. Neither of these deaths was attributed to the study drug.

Treatment with increasing doses of rhu-pGSN yielded a sustained dose-dependent increase in pGSN levels without much accumulation during 3 days of therapy. The half-life of rhu-pGSN was estimated to exceed 17 h, but sampling was limited to 24-h postdose measurements. These data support once-daily dosing in this population of moderately ill hospitalized CAP patients. A shorter dosing interval may be needed in more severely ill ICU patients, where pGSN is likely consumed more rapidly than in less severe infection.

Rhu-pGSN is structurally identical to endogenous human pGSN. Accordingly, treatment with rhu-pGSN was not expected to result in the formation of rhu-pGSN antibodies at day 28 of the study. However, we found that three individuals were positive for ADAs, all of whom were rhu-pGSN recipients in the multidose cohort. Curiously, antibodies were detected in 2 rhu-pGSN recipients at baseline before injection but disappeared at the end of study. Work is in progress to develop a more sensitive and specific ADA assay.

In conclusion, once-daily i.v. administration of rhu-pGSN to patients hospitalized on general medical wards with modestly severe CAP appeared safe and well tolerated in this small trial. Even with supraphysiological levels throughout the dosing interval, neither serious nor drug-related adverse events were observed in rhu-pGSN recipients given three consecutive days of therapy. A preliminary ADA assay detected putative anti-rhu-pGSN antibodies at day 28 in 3 rhu-pGSN recipients without recognized consequences. A larger proof-of-concept trial to confirm the safety and assess the efficacy of rhu-pGSN for the target population of ICU patients with sCAP is planned. As the world is besieged by COVID-19, rhu-pGSN should be viewed as a potential novel therapeutic agent that could save lives and reduce lung injury.

MATERIALS AND METHODS

Design and objectives.

We conducted a blinded, placebo-controlled, dose-escalation study from 28 August 2018 to 02 April 2019 to explore the safety and PK of rhu-pGSN added to standard of care in subjects hospitalized initially outside an ICU for treatment of acute CAP during the winter seasons in Australia and the Republic of Georgia (see Text S1 in the supplemental material). Approval was obtained from the Independent Ethics Committee for the conduct of the study at named sites, the protocol, informed consent documents, and any other written information that was provided to the subjects and any advertisements that were used. Written approval was obtained prior to recruitment of subjects into the study and shipment of study intervention.

The primary objective of this multicentered study was to assess the safety and tolerability of single and multiple doses of rhu-pGSN administered intravenously (i.v.) to hospitalized CAP patients admitted to non-ICU beds. Patients presenting with sCAP requiring immediate intensive care were purposely excluded so as not to confound assessment of drug-induced toxicity with disease morbidity. Secondary objectives were to determine the PK profile of i.v. rhu-pGSN. Exploratory objectives investigated the relationship among pGSN concentrations, levels of inflammatory markers, and clinical course. The immunogenicity of rhu-pGSN therapy was examined by testing for the presence of anti-pGSN antibodies on day 28. The study was neither designed nor powered to examine efficacy.

Participant selection.

Screening occurred within 24 h of presentation to the hospital. Male or female patients ≥18 years of age were eligible as long as the prospective subject could provide written informed consent, the duration of illness precipitating hospitalization was <14 days by history, the planned or actual admission to the hospital for CAP was to take place within 24 h of presentation to the hospital, and a primary admitting diagnosis of pneumonia was supported by a compatible clinical presentation with a documented infiltrate consistent with pneumonia visualized by chest radiography or chest tomography. Study treatment had to begin within 36 h of hospital presentation.

Patients were excluded if pregnant or lactating, on mechanical ventilation or vasopressor support, or directly admitted to an ICU bed. Administration of intravenous fluids to correct volume depletion with associated hypotension was permitted at the discretion of the treating physician. Patients who had used any investigational drug, were hospitalized in the preceding 30 days, or resided in a long-term care facility, where they remained persistently unable to participate in routine activities of daily living, were ineligible. Patients having active underlying cancer treated with systemic chemotherapy or radiation therapy during the last 30 days or with a known or suspected immunosuppressive disease or undergoing immunosuppressive therapy were excluded. Other exclusion criteria were congestive heart failure, myocardial infarction, pulmonary embolism, or cardiopulmonary arrest in the previous 30 days, weight of >100 kg, or otherwise unsuitable for study participation in the opinion of the investigator.

After screening, prospective subjects were randomized in a 3:1 ratio to receive rhu-pGSN or placebo, respectively. Treatment allocation was randomly assigned through an interactive web response system to a pharmacist who was unblinded to treatment assignment. The subject, caregivers, investigators, and sponsor were kept blinded. The duration of follow-up was 28 days after the first dose.

Study treatment.

Rhu-pGSN (BioAegis Therapeutics, North Brunswick, NJ) was produced and purified from Escherichia coli. Sites were provided with 10-ml glass vials containing lyophilized powder for reconstitution with 4.56 ml of sterile water to yield a 5-ml solution at a final concentration of 40 mg/ml rhu-pGSN in a proprietary stabilizing buffer. Dosing was based on actual body weight.

In the single-dose cohort, subjects received 1 dose of rhu-pGSN (or an equal volume of saline placebo) on day 1 as an i.v. infusion of 6 mg/kg at a rate between 5 and 20 ml/min through a standard 0.2-μm filter. In the 3 sequential multidose cohorts, subjects received once-daily doses of rhu-pGSN (or an equal volume of placebo) by i.v. infusion of 6, 12, or 24 mg/kg at a rate between 5 and 20 ml/min through a standard 0.2-μm filter for 3 consecutive days approximately 24 h (±60 min) apart. Masking was accomplished because the placebo solution was virtually indistinguishable from the active product.

A data and safety monitoring board (DSMB) was established to oversee the ongoing safety of subjects and to review the safety data from each dose cohort prior to dose escalation. The DSMB was comprised of 4 members: an independent expert in CAP (who acted as the DSMB chairperson), the sponsor's chief medical officer, the medical monitor of the contract research organization, and an unblinded statistician who was a nonvoting member. Dosing of rhu-pGSN could have been stopped and/or unblinded data requested at any time during the study at the discretion of the DSMB chair. Dose escalation occurred only after posttherapy safety information in all subjects in the prior cohort(s) had been reviewed on/after day 7 for the preceding single-dose and multidose cohorts.

Outcome measures.

The primary outcomes were the incidence, causality, and severity of AEs and SAEs. Causality was assigned by the site investigators as definitely, probably, or possibly drug related or definitely or probably unrelated. Severity was graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.03 (grade 1, mild; grade 2, moderate; grade 3, severe but not life-threatening; grade 4, life-threatening; grade 5, death related to AE) (46). Clinically significant laboratory and/or electrocardiographic abnormalities, death irrespective of cause, cause of death per investigator, relationship to study intervention, use of vasopressors, intubation, and transfer to an ICU were recorded.

Secondary outcome measures were PK parameters, including C_max (maximum observed plasma concentration), T_max (time to maximum observed plasma concentration), t_1/2 (elimination half-life), AUC_0–8 (area under the plasma concentration-time curve from time zero to 8 h), and AUC_0–t (area under the plasma concentration-time curve from time zero to 24 h). Specimens for pGSN levels were obtained within 30 min predose and then 5 to 10 min and 2, 8, 12, and/or 16 and 24 h (±30 min) after the end of the administration of each dose (although day 3 sampling in multidose cohorts was optional).

Exploratory outcome measures were baseline and sequential pGSN levels during the study, 28-day survival, ICU days, days on a ventilator or vasopressors, and duration of hospitalization. Severity indices at baseline and changes from baseline to day 28 were based on SOFA scores (sequential organ failure assessment; ranges from 0 [normal] to 24 [severe organ damage/dysfunction]) (47), PSI scores (pneumonia severity index; classifies patients according to risk classes I to V, from lowest to highest risk) (48), and CURB-65 scores (confusion, urea of >7 mmol/liter, respiratory rate of ≥30 breaths/min, blood pressure systolic <90 mm Hg or diastolic ≤60 mm Hg, and age of ≥65 years; evaluates risk of death and ranges from 0 [lowest] to 5 [highest]) (48, 49). Selected inflammatory biomarkers were monitored until the end of the study. Microbial etiologies were determined by customary work-up at each site.

Specialized assays.

Plasma gelsolin levels were measured using a sandwich enzyme-linked immunosorbent assay (ELISA). A rabbit polyclonal antibody against the 16 amino acids ATASRGASQAGAPQGC at the N-terminal region of pGSN (BioAegis Therapeutics Inc., North Brunswick, NJ) was immobilized on an ELISA plate to capture plasma gelsolin in plasma samples or the recombinant rhu-pGSN, used as a standard. The sample or the standard were combined with monoclonal mouse anti-gelsolin antibody clone 2C4 (Millipore-Sigma, St. Louis, MO, USA) and a horseradish peroxidase-conjugated anti-mouse IgG. The measured absorbance was directly proportional to the amount of gelsolin.

Inflammatory biomarkers were analyzed at Sonic Healthcare Limited (Sydney, New South Wales, Australia) using commercial kits for measuring interleukin-6 (IL-6) and procalcitonin (electrochemiluminescence immunoassay), high-sensitivity C-reactive protein (hsCRP; immunoturbidimetry) (Roche, Basel, Switzerland), tumor necrosis factor-α, transforming growth factor-β, IL-1β, IL-1ra, IL-2, IL-4, IL-10, and IL-17a (ELISA; R&D Systems, Minneapolis, MN, USA).

The formation of ADAs against rhu-pGSN at day 28 was determined using an affinity capture elution immunoassay that employed electrochemiluminescence for detection and was developed by TetraQ ADME Bioanalytics (Brisbane, Queensland, Australia). The mouse monoclonal anti-pGSN antibody clone 2C4, described above, was used as a positive control. A screening cut point was determined from the signal-to-noise ratio (S/N) of human serum samples from treatment-naive subjects. A response was considered potentially positive for ADAs if the S/N was above the screening cut point. The confirmatory threshold was determined from the change in signal in serum samples from treatment-naive subjects with or without preincubation with rhu-pGSN.

Prespecified statistical approach and analyses.

The sample size was intended to be 32 subjects in total, consisting of 6 subjects in the rhu-pGSN group and 2 subjects in the placebo group in each of the 4 dosing cohorts. Subjects missing doses or discontinuing at random (noninformative) before the primary visit on day 7 were replaced as originally randomized to receive rhu-pGSN or placebo. Subjects who discontinued the study after the day 7 visit were not replaced. No formal hypothesis testing was planned.

The selected sample size afforded a 90% probability of observing at least 1 AE of a certain type in 6 subjects at each rhu-pGSN dose level if the true underlying AE rate was 32%, in 8 subjects in a pooled placebo group if the true underlying AE rate was 21%, and in 24 subjects in a pool of all subjects who received rhu-pGSN if the true underlying AE rate was 10%. If no AEs of a certain type were observed, one could be 90% confident that the true underlying rate for that AE was at most the rate indicated above. All disposition, demographic, and pharmacodynamic analyses were based on the intention-to-treat (ITT) population, which included all randomized subjects according to their randomized treatment allocation. There were no cross-treatment errors, so the randomized and actual treatments were the same for all subjects. All safety analyses were based on the safety population, which included all subjects who received at least 1 dose of study intervention based on the actual treatment received. Because every participant received a dose of the assigned treatment, the safety and ITT population were identical. The PK population comprised all subjects in the safety population with adequate PK data to calculate PK parameters.

Data were summarized using counts and percentages of subjects for categorical variables, whereas means, standard deviations, geometric means, geometric standard deviations, medians, and ranges were computed for continuous variables. Plasma levels of pGSN were used to calculate PK parameters. Time to death or discharge was analyzed using Kaplan-Meier curves. There were no imputations or substitution for missing data points, except for the PK analyses, where pGSN concentrations below the quantitation limit were set to 0 if before the first quantifiable concentration and to one-half the lower level of quantification elsewhere for calculation of summary statistics at each time point.