Senolytics dasatinib and quercetin in idiopathic pulmonary fibrosis: results of a phase I, single-blind, single-center, randomized, placebo-controlled pilot trial on feasibility and tolerability

Summary

Background

Idiopathic pulmonary fibrosis (IPF) is an age-related, chronic, irreversible fibrotic lung disease. IPF is associated with increased senescent cells burden, which may be alleviated with administration of senescent cell targeting drugs termed ‘senolytics’. We previously conducted an open-label single-arm pilot study of the senolytic combination of dasatinib and quercetin (D + Q) in patients with IPF but lack of control group limited interpretation and next-stage trial planning. The primary objective of this confirmatory randomized placebo-controlled pilot trial (RCT; NCT02874989) was to report adverse events with D + Q and inform study feasibility for future efficacy trials.

Methods

Twelve participants with IPF aged >50 years were blinded and randomized at a 1:1 ratio to either receive three weeks of D + Q (D: 100 mg/d and Q: 1250 mg/d, three consecutive days per week) or matching placebo.

Findings

All participants completed the scheduled drug dosing regimen (108/108 doses) and planned assessments (60/60). While the placebo arm reported fewer overall non-serious AEs (65 vs 22), there were no serious adverse events related to D + Q. Most AEs in the D + Q arm are common in IPF patients or anticipated side effects of D. Sleep disturbances and anxiety were disproportionately represented in the D + Q arm (4/6 vs 0/6). Frailty, pulmonary, or physical function were explored before and after intermittent D + Q; though under-powered to evaluate change, these measures do not appear to differ meaningfully between groups.

Interpretation

Intermittently-dosed D + Q in patients with IPF is feasible and generally well-tolerated. Further prospective studies, such as a larger RCT, are needed to confirm the safety and efficacy of D + Q in patients with IPF.

Funding

This work was supported by National Institutes of Health grants R33AG61456 (JLK, TT), Robert and Arlene Kogod (JLK, TT), the Connor Fund (JLK, TT), Robert J. and Theresa W. Ryan (JLK, TT), and the Noaber Foundation (JLK, TT) San Antonio Claude D. Pepper Older Americans Independence Center’s (OAIC)Pilot/Exploratory Studies Core (PESC) Grant (AMN, NM); NIHK01 AG059837 (JNJ), P30 AG021332 (SBK, JNJ); NIHR37 AG013925 (JLK), the Connor Group (JLK), Glenn/AFAR BIG Award (JLK), Robert J. and Theresa W. Ryan (JLK), and the Noaber and Ted Nash Long Life Foundations (JLK).

Research in context.

Evidence before this study

Idiopathic pulmonary fibrosis (IPF) is a quintessential age-related disease, where lung fibroblasts and epithelial cells become prematurely senescent. Senolytics have demonstrated significant improvement in lung compliance and exercise capacity in pre-clinical models. In 2019, we published an open-label pilot study of the senolytic combination of dasatinib and quercetin (D + Q) in fourteen IPF patients.

Added value of this study

Here, we report the second stage of the pilot trial (NCT02874989) with this small, single-blinded, single-center, randomized, placebo-controlled trial of intermittently-dosed D + Q.

Implications of all the available evidence

This D + Q regimen is feasible and well-tolerated in IPF patients, but this study suggests important areas for safety and symptom monitoring in next-stage trials.

Introduction

Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive, and highly morbid disease with a median survival of 3.8 years and limited treatment options.¹ IPF is characterized physiologically by a restrictive ventilatory defect and compromised gas exchange leading to progressive dyspnea, impaired quality of life, and ultimately death.^2,3 Currently, there are two FDA-approved antifibrotics for IPF, nintedanib and pirfenidone, that have been shown to only slow disease progression but not halt or reverse progressive fibrosis.

IPF pathophysiology is characterized by exaggerated manifestations of the hallmarks of aging in alveolar epithelium and fibroblasts, including cellular senescence.4, 5, 6 The defining characteristics of cellular senescence are growth arrest, resistance to apoptosis, stem cell exhaustion and the production of pro-inflammatory and pro-fibrotic senescence-associated secretory phenotype (SASP) factors in response to cellular stress.7, 8, 9 Preclinical evidence suggests that senescent fibroblasts and alveolar epithelial cells contribute to pulmonary fibrosis and supports cellular senescence as a potential target for treatment for bleomycin-induced lung injury model of pulmonary fibrosis that may be translationally relevant to humans.^5,10,11

Senotherapeutic treatments that target cellular senescence have been accomplished in animal models and demonstrated improved and functional status.^12,13 Senolytics selectively eliminate senescent cells by inhibiting senescent cell anti-apoptotic pathways (SCAPs). The synergistic combination of dasatinib and quercetin (D + Q) were the first senolytic agents identified using a hypothesis-driven drug discovery approach led by Zhu et al. in 2015.¹⁴ Dasatinib is an inhibitor of multiple tyrosine kinases with mechanism of action linked to a broad targeting of Src kinases, and is currently indicated for use as a second line chemotherapeutic agent for treatment of chronic myeloid leukemia resistant to another tyrosine kinase inhibitor, imatinib.^15,16 Quercetin is a natural product and nonspecific kinase inhibitor that targets PI3K/AKT pathway modules, as well as BCL-2, insulin/IGF-1, and HIF-1α SCAP networks components, and is senolytic, possibly as a consequence of its inhibitory effects on multiple SCAP genes (i.e. PI3K and other kinases).^14,17,18 Moreover, D + Q have a short half-life of ∼4-hr and 12-h elimination half-lives of D and Q, respectively, and an intermittent or “hit-and-run” dosing approach can be effective. In bleomycin-induced pulmonary fibrosis models, monotherapy with quercetin markedly reduced senescent markers.¹⁰ The combination of D + Q is complementary, yielding increased senolysis when co-administered, and decreases senescent cell burden and SASP in human tissues within 48 h of administration.^14,19 D + Q alleviates a range of age- and senescence-related disorders in mice,^5,12,13,20, 21, 22, 23, 24 and has been found to significantly improve lung function and exercise capacity with intermittent dosing at onset of bleomycin lung-injury.⁵

Recently, we translated these preclinical findings to humans in a single-arm open label proof of concept study of D + Q in patients with stable IPF. We reported favorable study design feasibility, acceptable safety profile, and noted a possible improvement in physical function, though lack of control group limits interpretation and use for future trial planning.²⁵ The primary objective of this confirmatory randomized, placebo-controlled trial of D + Q in patients with IPF was to evaluate the feasibility and tolerability of D + Q relative to placebo to facilitate the design of larger efficacy trials. Our primary outcome is feasibility and tolerability as assessed by: (1) severe adverse event linked to drug administration; (2) unanticipated adverse event more than moderate; (3) proportion taking at least 80% of intended dose (7 of 9 courses). Exploratory measures of pulmonary and physical function as well as respiratory symptoms and fatigue were included to examine measurement characteristics that may aid future trial planning.

Methods

Participants

The study (NCT02874989) was conducted at the University of Texas Health San Antonio (UTHSA) under the auspices of the ethics committee (IRB reference HSC20170199H) with informed consent from participants between December 2018 and May 2019. Eligible participants were recruited from the UTHSA IPF clinic and were invited to participate if they met pre-specified inclusion and exclusion criteria. Included participants required confirmed IPF (by ATS/ERS guidelines with high-resolution computed tomography (HRCT) and/or lung biopsy), clinical stability for the prior year (≤2 acute exacerbation of IPF), and stable antifibrotic therapy for the prior three months. Participants were screened for exclusion criteria and deemed ineligible if they had known or suspected clinically significant pulmonary hypertension (either pre-, post-, or mixed pre- and post-capillary) or cor pulmonale by echocardiography or cardiac catheterization, lung transplant, myocardial infarction, angina, hospitalization for cardiac etiology, stroke or transient ischemic attack in the past 6 months, chronic heart failure, liver disease, neurologic condition, illicit drug use within 5 years, QTc prolongation (>450 ms), pancytopenia, glomerular filtration rate (GFR) < 30 mL/min/1.73 m² or were taking QTc prolonging anti-arrhythmic, anti-platelet or anti-coagulant medications as well as quinolone antibiotics or medications with similar metabolism to D + Q.²⁵

Trial design and randomization

This prospective study as outlined in Fig. 1 was designed to inform study feasibility for future efficacy trials, assess potential nonadherence and identify potential safety concerns. There were no changes to eligibility criteria during enrollment. Participants were masked to treatment group status during randomization and throughout the treatment regimen. Sample size was selected based on feasibility. Block randomization intentionally did not stratify by factors such as age, sex and IPF severity, given the small sample size. UTHSA enrolled participants and assigned participants to intervention arms in a single-blind fashion by M.A. Biaglow, PharmD. Study coordinators called participants on a weekly basis to monitor for adverse symptoms. Laboratory adverse events were assessed within two weeks of completing D + Q.The data safety monitoring board (DSMB) was hosted by Wake Forest School of Medicine and convened every 6 months. Stephen Kritchevsky, Ph.D was the medical safety officer. The DSMB did not request revisions.

Fig. 1.

Schematic of study design. D + Q (100mg/1250 mg) or placebo, self-administered daily for three consecutive days on three consecutive weeks. Baseline and post-treatment assessments included PFTs, physical function, clinical chemistries, and subjective respiratory health and fatigue scales. Adherence and AEs were reported within the 24 h following dosing days. Participants were followed to 45 days for any clinical complications but not actively evaluated following completion of post-treatment assessment.

Interventions

Dose and administration of D + Q was identical to that performed in the previously published open label study, which was based on preclinical and pharmacokinetic studies.^5,21,26 Participants self-administered either placebo or intermittent D + Q, D: 100 mg orally per day; Sprycel, Bristol Myers Squibb and Q: 1250 mg orally per day (quercetin phytosome Thorne Research Sophora japonica concentrate [leaf]/phosphatidylcholine complex from Sunflower. Doses were taken over three consecutive days in three consecutive weeks, resulting in nine total doses (Fig. 2). Nintedanib was withheld on dosing days only with D + Q administration to avoid potentially deleterious effects of dual tyrosine kinase inhibition. Pirfenidone was continued throughout the trial.

Fig. 2.

CONSORT flow diagram. Twelve IPF patients with mild to moderate disease, controlled comorbidities and clinical stability were recruited from UTHSA then 1:1 single-blinded randomized to either D + Q or placebo. All twelve participants completed the regimen, and analyses maintained all participants in initial randomized groups.

Trial assessments

Serial adverse events (AE) and adherence assessments were performed approximately 24 h after completion of all three treatment courses and queried common AEs with chemotherapy and IPF medications. The severity of symptoms was graded according to the Medical Dictionary for Regulatory Activities (mild = tolerable, moderate = interferes with normal activity, severe = precludes normal activity). AE surveillance included fasting clinical chemistries and complete blood count prior to treatment and within one week of completion.

Objective pulmonary and physical function, as well as frailty and subjective health assessments, were completed at baseline and within one week after the intervention phase. The pulmonary function included forced expiratory volume in 1 s (FEV1) and forced vital capacity (FVC), and physical function included 6-min walk distance (6MWD), 4-m gait speed, timed chair-stands, Short Physical Performance Battery (SPPB), and grip strength as described in our prior study.²⁵ Percent predicted FEV1 and FVC were based on age, ethnicity, gender and height. The frailty index (FI-LAB) was calculated based on 34 clinical chemistries. Participant fatigue and respiratory symptoms were assessed with the Fatigue Severity Scale (FSS), Pittsburgh Fatigability Scale (PFS), Modified Medical Research Council (mMRC) dyspnea scale, St George Respiratory Questionnaire (SGRQ), and UCSD shortness of breath questionnaire (UCSD-SoBQ). AE surveillance included fasting clinical chemistries and complete blood count prior to treatment and within 1 week of completion. There were no changes to trial outcomes after the trial commenced and no allowance for multiplicity across outcomes.

Power and sample size

The sample size of n = 6 per arm (D + Q and placebo arms, 12 participants randomized in total) was selected to assess feasibility and tolerability rather than efficacy. The goal enrollment was 12 (26 total in combination with our previously published open-label pilot).

Statistical analysis

Safety and tolerability were assessed by the frequency and severity of AEs in both arms. The AEs were graded either as mild, moderate, or severe as above, and the treatment arms were compared with Fisher exact test. The pre/post difference score was computed for clinical outcomes (pulmonary function, physical function, frailty, respiratory health), and the 95% confidence intervals were computed using the one-sample (within group) or two-sample (between group) t-test statistics. All analyses were performed using an accountable data analysis process in R v4.0 (Vienna, Austria). REDCap data management systems were used for all data entry and handling; group assignments and randomization scheme were not included in database.

Role of funders

This study was funded by National Institutes of Health, and supporting funders were not involved in study design, data collection, data analyses, interpretation, or writing of report.

Results

A total of 12 participants met the inclusion criteria, consented, and randomized in a 1:1 fashion to either D + Q (n = 6) or matching placebo (n = 6) (Fig. 2). All invited participants were followed by UTHSA study clinicians, thus only participants that met the previously published inclusion criteria were screened by the study coordinator. Gender was self-reported by participants. There were no premature withdrawals, and endpoints of retention (12/12 participants, 100%), adherence (108/108 doses, 100%), and clinical assessment completion (12/12, 100%) were very high. The baseline demographics and clinical characteristics are summarized in Table 1.

Table 1.

Patient baseline characteristics.

	DQ	Placebo
N	6	6
Age (mean ± SD)	69.2 ± 7.5	64.8 ± 7.0
Male sex (n (%))	5 (83)	5 (83)
MOCA (mean ± SD)	27.0 ± 2.6	27.5 ± 2.7
Non-Hispanic White (n (%))	5 (83)	3 (50)
Hispanic (n (%))	1 (17)	3 (50)
Active tobacco use	0	0
Physiologic parameters
FEV1 (%predicted)	69.8 ± 9.0	85.2 ± 16.4
FVC (%predicted)	61.0 ± 6.7	75.5 ± 12.2
IPF severity by FVC (n (%))
Preserved (>90%)	0 (0.0)	1 (17)
Mild (80–90%)	0 (0.0)	1 (17)
Moderate (50–80%)	6 (100)	4 (67)
Physical function (mean ± SD)
6-min walk distance (m)	398.8 ± 89.0	420.9 ± 73.7
4-m gait speed (m/s)	1.0 ± 0.3	1.2 ± 0.3
Timed chair-stands (s)	15.2 ± 3.1	11.2 ± 3.5
SPPB scorea	9.7 ± 1.4	11.3 ± 0.8
Grip strength (kg)	35.3 ± 10.4	32.1 ± 9.2
Medications
Antifibrotic medications (n (%))
None	3 (50)	1 (17)
Nintedanib	2 (33)	3 (50)
Pirfenidone	1 (17)	2 (33)
Non-IPF medications (n (%))
≤ 5	0	2 (33)
6–10	3 (50)	2 (33)
11–15	1 (17)	1 (17)
≥ 16	2 (33)	1 (17)
Comorbid conditions (n (%))
0–3	2 (33)	5 (83)
4–6	3 (50)	1 (17)
> 6	1 (17)	0
Allergic rhinitis (%)	5 (83)	2 (33)
Hypertension (%)	2 (33)	1 (17)
Colon polyps (%)	2 (33)	1 (17)
Trouble sleeping (%)	4 (66.7)	1 (16.7)

Montreal Cognitive Assessment: ≥27: healthy; ≤22: significant cognitive impairment.

^aShort Physical Performance Battery (SPPB).

The AEs are shown in Table 2, which corroborates the morbid natural history of IPF as 50% of both arms reported severe AEs. Overall, there were more adverse events reported in the D + Q arm (65 vs 22). The most frequent non-serious AEs in the D + Q arm, were feeling unwell (83%, 5/6) and cough, nausea, fatigue, weakness, and headache (67%, 4/6). Nausea, weakness, headache, and sleep disturbance were disproportionally reported in the D + Q arm. There was a serious AE after study completion (29 days after final dose D + Q), in the context of infectious and post-operative complications. After study completion, a D + Q arm patient with a history of chronic symptomatic cholelithiasis was hospitalized with sepsis from cholecystitis, requiring cholecystectomy, complicated by intra-operative respiratory failure that ultimately resulted in death. This serious AE occurred 59 days after D + Q and was adjudicated as not related to D + Q by the UTHSA IRB given the patient's increased risk for cholecystitis based on known history of cholelithiasis, increased risk of acute respiratory failure due to underlying IPF, and short half-life of D + Q (∼4 and 12-h elimination half-lives of D and Q). Importantly, there were no reports cases of myelosuppression, a common adverse event related to dasatinib, observed during this acute decompensation or in our study population at post-treatment follow-up (Supplemental Table S1).

Table 2.

Prevalence and severity of adverse events assessed within 24 h of treatment completion.

Adverse eventsb	D + Q				Placebo
	Overall	Mild	Mod	Severe	Overall	Mild	Mod	Severe
Patients experience AEs (n (%))		6 (100%)	5 (83%)	3 (50%)		3 (50%)	5 (83%)	3 (50%)
Non-serious AEs	65	21	19	25	22	8	10	4
Serious AEa	–	–	–	–	–	–	–	–
Fatal AEa	–	–	–	–	–	–	–	–
Adverse event leading to discontinuation	–	–	–	–	–	–	–	–
Respiratory adverse events
Cough	4	2	–	2	3	1	1	1
Shortness of breath	3	–	–	3	2	1	–	1
Nasal symptoms	2	–	1	1	1	–	1	–
Worsening IPF symptoms	2	–	1	1	–	–	–	–
Flu like symptoms	1	–	–	1	–	–	–	–
Bronchitis	1	–	–	1	–	–	–	–
General health & well-being
Unwell	5	1	2	2	1	1	–	–
Fatigue	4	–	1	3	2	1	1	–
Weakness	4	–	2	2	–	–	–	–
Sleep disturbance	4	3	–	1	–	–	–	–
Anxiety	3	1	1	1	–	–	–	–
Fever	2	1	1	–	–	–	–	–
Depression	2	–	1	1	1	–	1	–
Weight change	1	–	1	–	–	–	–	–
Gastrointestinal & appetite
Nausea	4	2	1	1	–	–	–	–
Other GI	3	1	2	–	2	–	–	2
Appetite change	3	1	1	1	–	–	–	–
Diarrhea	2	–	1	1	1	–	1	–
Vomiting	1	1	–	–	–	–	–	–
Constipation	1	1	–	–	2	–	2	–
Pain or discomfort
Headache	4	–	2	2	–	–	–	–
Pain	3	2	–	1	3	1	2	–
Numbness	1	1	–	–	1	1	–	–
Mouth, eyes
Sore throat	2	2	–	–	1	–	1	–
Dry/watery eyes	2	1	1	–	2	2	–	–
Smell	1	1	–	–	–	–	–	–

Compares rate of specific adverse events of any level of severity.

^aExperienced after completion of study intervention and assessments (day 42 post-DQ).

^bMild = tolerable, moderate = interferes with normal activity, severe = precludes normal activity.

The objective and subjective clinical outcomes are summarized in Table 3, Table 4, respectively. The study was not designed to demonstrate treatment by time interactions for any of these exploratory endpoints; as such results are presented graphically and as means and confidence intervals for use in future trial planning. However, change in FVC, FEV1, 6MWD, SPPB, fatigability do not appear to differ meaningfully between the treatment groups (Fig. 3). We did collect blood and urine that may allow future studies to develop biomarkers for senolytic trials, specifically, Glycoprotein nonmetastatic melanoma protein B (GPNMB or Osteoactivin) compared with placebo (Supplemental Table S1).

Table 3.

Changes in pulmonary, physical function, and frailty index.

	Baseline		Post		Difference
	Placebo	D + Q	Placebo	D + Q	Placebo	D + Q	95% CI
	n = 6	n = 6	n = 6	n = 6
Pulmonary functiona
FEV1 (L)	2.5 ± 0.5	2.1 ± 0.7	2.5 ± 0.5	2.2 ± 0.6	0.01 ± 0.2	0.1 ± 0.1	[−0.2, 0.1]
FEV1 (% predicted)	85.2 ± 16.4	69.8 ± 9.0	86.3 ± 16.6	72.0 ± 7.2	1.2 ± 5.3	2.2 ± 3.1	[−6.7, 4.7]
FVC (L)	2.9 ± 0.6	2.5 ± 0.8	2.9 ± 0.6	2.6 ± 0.7	0.1 ± 0.1	0.0 ± 0.1	[−0.1, 0.1]
FVC (% predicted)	75.5 ± 12.2	61.0 ± 6.7	77.2 ± 12.7	62.5 ± 6.2	1.7 ± 2.8	1.5 ± 2.4	[−3.2, 3.6]
Physical function
6-min walk distance (m)	420.9 ± 73.7	398.8 ± 89.0	437.5 ± 88.4	392.3 ± 77.6	16.6 ± 20.7	−6.4 ± 15.9	[−1, 46.9]
4-m gait speed (m/s)	1.2 ± 0.3	1.0 ± 0.3	1.3 ± 0.2	1.2 ± 0.1	0.1 ± 0.3	0.2 ± 0.3	[−0.4, 0.3]
Timed chair-stands (s)	11.2 ± 3.5	15.2 ± 3.1	9.8 ± 3.2	14.3 ± 3.0	−1.4 ± 2.7	−1.0 ± 2.3	[−3.7, 2.8]
SPPB scoreb	11.3 ± 0.8	9.7 ± 1.4	11.5 ± 0.8	10.5 ± 1.0	0.2 ± 1.0	0.8 ± 1.0	[−1.9, 0.6]
Grip strength (kg)	32.1 ± 9.2	35.3 ± 10.4	39.8 ± 7.9	37.1 ± 10.0	7.7 ± 6.6	1.8 ± 4.7	[−1.6, 13.4]
Frailty index
FI-LABc	0.10 ± 0.06	0.10 ± 0.05	0.12 ± 0.08	0.12 ± 0.1	0.02 ± 0.04	0.02 ± 0.05	[−0.1, 0.1]

- 95% confidence intervals compare baseline value and mean differences for placebo vs. D + Q groups.

^aRepresented as mean ± SD. FEV1: forced expiratory volume in 1 s (liters, L); FVC: forced vital capacity (liters, L); Percent predicted values based on age, sex, race, and height.

^bShort Physical Performance Battery is a combination of 4mGS, chair-stands, and balance (0–12 with 12 best).

^cFI-LAB (lab-based frailty index score derived from analytes in/out of reference range for 34 blood-based clinical chemistries).

Table 4.

Changes in subjective respiratory health related quality of life and fatigue.

	Baseline		Post		Difference
	Placebo	D + Q	Placebo	D + Q	Placebo	D + Q	95% CI
	n = 6	n = 6	n = 6	n = 6
Respiratory healtha
mMRC dyspnea	2.2 ± 1.0	2.2 ± 1.0	2.0 ± 1.1	2.5 ± 0.8	−0.2 ± 0.4	0.3 ± 1.4	[−1.9, 0.9]
SGRQ	36.0 ± 5.0	40.7 ± 8.6	35.0 ± 4.7	41.0 ± 8.0	−1.0 + 1.9	0.3 ± 2.3	[−4.1, 1.4]
UCSD-SoBQ	13.5 ± 8.7	44.3 ± 37.3	13.0 ± 9.3	40.2 ± 36.0	−0.5 + 6.2	−4.2 + 9	[−6.4, 13.8]
Perceived fatigue scalesb
Pittsburgh fatigability, physical	13.5 ± 8.1	24.7 ± 14.8	14.2 ± 8.2	22.8 ± 14.1	0.7 ± 5.1	−1.8 ± 3.7	[−3.2, 8.2]
Pittsburgh fatigability, mental	4.5 ± 7.7	18.3 ± 17.6	4.3 ± 7.9	11.7 ± 15.9	−0.2 ± 1.2	−6.7 ± 7.4	[−1.2, 14.2]
Fatigue severity	33.6 ± 18.8	39.7 ± 18.0	33.8 ± 22.5	36.5 ± 20.6	0.2 ± 7.2	−3.2 ± 5.0	[−4.8, 11.4]

- 95% confidence intervals compare baseline value and mean differences for placebo vs. D + Q groups.

^aRepresented as mean ± SD. mMRC: Modified Medical Research Council (0 = dyspnea with strenuous exercise, 4 = at rest). SGRQ: St George Respiratory Questionnaire (0 = no impairment, 100 = ADL impairment). UCSD-SoBQ: shortness of breath questionnaire (0 = no limitations, 120 = ADL impairment).

^bPFS-physical: ≥15 = higher fatigability. PFS-mental: ≥13: higher fatigability. Fatigue severity scale (9 = no limitations, 67 = ADL limitation).

Fig. 3.

Change plots of key clinical variables. Pre: within 1 month before treatment; post: 1 week after treatment; Red line: mean value (a) Placebo group (n = 6); (b) D + Q arm (n = 6); FVC: Forced vital capacity, 6MWD: 6-min walk distance. SPPB: Short Physical Performance Battery, scored 0–12 with 12 being the best performance.

Discussion

This pilot study of senolytics in IPF patients demonstrates that intermittent self-administered D + Q is tolerated and an acceptable approach for phase II RCT that assess the clinical risks and benefits. We demonstrated study design feasibility based on high participant adherence and visit compliance. D + Q was associated with a roughly 3-fold increase in non-serious AEs and modest increase in limiting symptoms; none of which resulted in premature treatment discontinuation. The most common non-serious AEs that occurred in the D + Q arm were consistent with known AEs of D (fatigue, nausea, headache, diarrhea, decreased appetite, and feeling unwell).²⁷ Burden of IPF-associated symptoms was balanced between arms (cough, shortness of breath, depression, GI symptoms [common side-effect of antifibrotics]) and was consistent with the previously reported open-label pilot study. The majority of these AEs was reported by a minority of participants. The trends toward increased sleep disturbances (4/6 vs. 0/6 placebo) and anxiety (3/6 vs. 0/6) are not commonly reported AEs in either hematologic malignancy or IPF patients in general. Thus, these non-serious AEs should be monitored in subsequent trials and additional follow-up is needed to fully evaluate persistent AEs.

The serious post-operative AE that occurred 29 days after study completion was unlikely related to D + Q as evidenced by the ∼4 and 12-h elimination half-lives of D and Q, respectively. Moreover, there are rare (0.1%–<1%) and non-existent occurrences of hepatobiliary disorders with D and Q, respectively. An infection has been reported in 13% of patients on D but is usually mild and often correlated with dose-dependent myelosuppression (29–79% with moderate neutropenia [ANC<1000/μL]), which was not observed in this RCT or the pilot study.²⁸

Small sample size is the main limitation to this pilot trial. The purpose of this trial was not to determine efficacy of D + Q on primary clinical outcomes relevant to IPF, such as FVC or 6MWD. Thus, this study has insufficient power for efficacy conclusions, and sample size precluded subgroup analysis or over-interpreting results from pilot study. Furthermore, there were fewer patients on antifibrotic medications in the D + Q arm: 3/6 (50%) vs 5/6 (83%) in the placebo arm. Antifibrotic medications can modulate downstream extra-cellular matrix remodeling.^29,30 Thus, they may have synergistic effects with clearance of pro-fibrotic senescent fibroblasts by D + Q and could be factored into subsequent randomization protocols. This criteria and group balance should be carefully considered in future trials. Additionally, there are inherent limitations to our assessments of function. For example, 6MWD assesses the integrated function of cardiac, pulmonary and musculoskeletal systems, which are subject to learning effects and patient effort and can vary considerably, especially in patients with pulmonary disease and higher 6MWD.³¹ Moreover, global measures of patient function like 6MWD are susceptible to learning effect with repeated trials. Additionally, our study duration was only three weeks, which is insufficient time to allow for clinically detectable changes in pulmonary function or fibrosis; future trials will require large sample size and longer follow-up to measure meaningful changes in physical and pulmonary function and patient reported outcomes such as pulmonary-related quality of life.

Identifying senescence biomarkers and their relationship with clinical IPF severity as well as change during senolytic therapy will be valuable for gauging senescent cell destruction and further pathophysiological understanding of IPF. Unfortunately, there are no current clinically validated circulating biomarkers of cellular senescence. A recent post hoc analysis using samples obtained before and after D + Q treatment in IPF patients suggests a potential improvement of urinary α-Klotho, which is geroprotective across multiple age-related disease.³² In the present study, GPNMB, an endogenous glycoprotein related to inflammation and cell senescence, may be reduced in D + Q group; future studies exploring effect of senolytics on change in biomarkers should consider including this marker in biomarker panels. Future trials assessing serologic panels of senescent biomarkers will be critical in the clinical translation of senolytics and direct treatment by identifying suboptimal senolytic responses and possibly prompt additional treatment. Additionally, future studies with D + Q are recommended to monitor glucose-lipid biomarkers. Though we noted potentially elevated levels of fasting glucose and LDL cholesterol in D + Q compared with placebo, previous studies in chronic myeloid leukemia noted lower blood glucose levels and amelioration of diabetic hyperglycemia.33, 34, 35, 36

These data will aid in the study design of future larger prospective RCTs of D + Q in patients with IPF and can inform clinical trials in other age-related diseases, such as osteoporosis, Alzheimer's dementia, chronic kidney disease, and frailty in cancer survivors.^14,21,37 Importantly, D + Q in patients with IPF requires further investigation to establish safety and efficacy prior to clinical use, and nonserious adverse events and potential treatment-related symptoms must be closely monitored during and after D + Q administration.

Contributors

Nambiar AM, Kellogg DL III, Justice JN, Goros M, Gelfond J, Pascual R, Hashmi SK, Masternak MM, Prata L, Kritchevsky SB, LeBrasseur NK, Limper AH, Tchkonia T, Musi N, Kirkland JL

JK, AN, JJ, NM, SK, and NL conceived and planned the experiments. AN and NM carried out the experiments. TT and MM contributed to sample preparation and assays. MG and JG curated data and perform statistical analysis. AN, DK, JJ, JK, NM, SK, NL, RP, AL, and TT contributed to the interpretation of the results. DLK took the lead in writing the manuscript with JJ, NL, SH, RP, MM, AN, SK, TT, JK, LP. contributing to review and editing. All authors provided critical feedback helped shape the research, analysis and manuscript and approved the final version.

Data sharing statement

De-identified data supporting this study are available on request; please contact our corresponding author.

Declaration of interests

TT and JK have a financial interest related to this research including patents and pending patents on senolytic drugs, and their uses are held by Mayo Clinic. This research has been reviewed by the Mayo Clinic Conflict of Interest Review Board and was conducted in compliance with Mayo Clinic Conflict of Interest policies.