Machine learning prediction on number of patients due to conjunctivitis based on air pollutants: a preliminary study

OBJECTIVE

A prediction of the number of patients with conjunctivitis plays an important role in providing adequate treatment at the hospital, but such accurate predictive model currently does not exist. The current study sought to use machine learning (ML) prediction based on past patient for conjunctivitis and several air pollutants. The optimal machine learning prediction model was selected to predict conjunctivitis-related number patients.

PATIENTS AND METHODS

The average daily air pollutants concentrations (CO, O3, NO2, SO2, PM10, PM2.5) and weather data (highest and lowest temperature) were collected. Data were randomly divided into training dataset and test dataset, and normalized mean square error (NMSE) was calculated by 10 fold cross validation, comparing between the ability of seven ML methods to predict the number of patients due to conjunctivitis (Lasso penalized linear model, Decision tree, Boosting regression, Bagging regression, Random forest, Support vector, and Neural network). According to the accuracy of impact prediction, the important air and weather factors that affect conjunctivitis were identified.

RESULTS

A total of 84,977 cases to treat conjunctivitis were obtained from the ophthalmology center of the Affiliated Hospital of Hangzhou Normal University. For all patients together, the NMSE of the different methods were as follows: Lasso penalized linear regression: 0.755, Decision tree: 0.710, Boosting regression: 0.616, Bagging regression: 0.615, Random forest: 0.392, Support vectors: 0.688, and Neural network: 0.476. Further analyses, stratified by gender and age at diagnosis, supported Random forest as being superior to others ML methods. The main factors affecting conjunctivitis were: O3, NO2, SO2 and air temperature.

CONCLUSIONS

Machine learning algorithm can predict the number of patients due to conjunctivitis, among which, the Random forest algorithm had the highest accuracy. Machine learning algorithm could provide accurate information for hospitals dealing with conjunctivitis caused by air factors.

OP0126 INFECTIONS AND SERIOUS INFECTIONS IN THE FILGOTINIB RHEUMATOID ARTHRITIS PROGRAM

Background:

The Janus kinase (JAK)-1 preferential inhibitor filgotinib (FIL) improved rheumatoid arthritis (RA) signs and symptoms in 3 phase (P)3 trials.1–3 Like other RA therapies, JAK inhibition is associated with increased infection rates.4

Objectives:

To assess long-term safety across the FIL program regarding infections, including serious infections (SI).

Methods:

Patients (pts) meeting 2010 ACR/EULAR RA criteria in pooled analysis of P2 DARWIN 1–2 (D1–2), P3 FINCH 1–3 (F1–3), and long-term extension studies (DARWIN 3, FINCH 4) were included. The placebo (PBO)-controlled as-randomised data set included pts receiving FIL 100 mg (FIL100), FIL 200 mg (FIL200), or PBO up to week (W)12 (D1–2, F1–2). The active-controlled as-randomised data set included pts receiving FIL100, FIL200, adalimumab (ADA), or methotrexate (MTX) up to W52 (F1, F3). The long-term as-treated data set included pts in all 7 studies receiving FIL100 or FIL200; data after rerandomisation were included and contributed to treatment received.

Exposure-adjusted incidence rates (EAIRs) per 100 patient-years exposure (PYE) and differences with 95% confidence intervals (CIs) were calculated using Poisson regression; EAIRs for tuberculosis (TB) in active controlled sets were calculated using an Exact Poisson method. Kaplan-Meier (KM) event probabilities with 95% CIs were provided for SI. If pts had multiple events within the same treatment period, only the first event was counted in EAIR calculation; PYE were calculated up to the last follow-up time or day before next treatment, including after first event. For KM analysis, time to event was calculated until the first event.

Results:

Of 2267/1647 pts in as-treated set receiving FIL200/FIL100, 1697 had treatment-emergent infection; 118 were SI. Baseline potential risk factors for pts with SI are in Table.

Table 1.

Baseline characteristics of pts with/without treatment emergent SIa

Parameter, n (%)SIN = 92No SIN = 2491Medical history Chronic lung disease13 (14.1)125 (5.0) Chronic renal disease3 (3.3)23 (0.9) Infections and infestations29 (31.5)499 (20.0)Baseline body mass index, kg/m2 <3064 (69.6)1749 (70.2) ≥3028 (30.4)742 (29.8)Age, years <6567 (72.8)2006 (80.5) ≥6525 (27.2)485 (19.5)Former/current smoker30 (32.6)677 (27.2)Oral corticosteroids, mg <7.528 (56.0)731 (66.1) ≥7.522 (44.0)375 (33.9) Missing data421385

aPhase 3 (FINCH 1-4) studies, as randomised.

SI, serious infection.

In 12W PBO-controlled period, infection rates were 17.9%/15.6%/13.3% for FIL200/FIL100/PBO. In 52W ADA-controlled period, infection EAIRs (95% CIs)/100 PYE were 46.9 (40.9, 53.7)/43.7 (38.0, 50.4)/43.4 (36.5, 51.5), FIL200/FIL100/ADA; and 38.5 (33.8, 43.9)/39.0 (31.1, 48.8)/42.2 (36.1, 49.3), FIL200/FIL100/MTX in 52W MTX-controlled period; 24.8 (23.1, 26.5)/34.4 (30.4, 38.8), FIL200/FIL100 in long-term analysis. In 12W PBO-controlled period, there was no active TB for FIL200/FIL100/PBO. In 52W ADA-controlled period, active TB EAIRs (95% CIs)/100 PYE were: 0 (0.0, 0.8)/0 (0.0, 0.8)/0.3 (0.0, 1.9), FIL200/FIL100/ADA and 0 (0.0, 0.6)/0 (0.0, 1.9)/0 (0.0, 1.0), FIL200/FIL100/MTX in 52W MTX-controlled period; 0/0.1 (0.0, 0.5), FIL200/FIL100 in long-term analysis.

SI rate or EAIRs are in Figure. Most common infections were upper respiratory tract infection and nasopharyngitis; majority were low grade. Pneumonia was most common SI (<1%). In long-term population, event probability (95% CI) of SI was 2.2% (1.6, 2.9)/2.5% (1.8, 3.4) for FIL200/FIL100 at 52W. In F1–3 (excluding data after rerandomisation), there were no significant changes in mean neutrophil and lymphocyte counts; values remained within normal limits up to W52 for all arms.

Conclusion:

EAIRs of infections and SI for FIL were similar to PBO, ADA, and MTX. At 52W, incidence rates of SI were comparable for FIL100 and FIL200. Long-term SI EAIR for FIL100 was slightly higher than for FIL200.

References:

[1]Genovese et al. JAMA. 2019;322:315–25.

[2]Westhovens et al. Ann Rheum Dis. 2021; online first.

[3]Combe et al. Ann Rheum Dis. 2021; online first.

[4]Strand et al. Arthritis Res Ther. 2015;17:362.

Disclosure of Interests:

James Galloway Speakers bureau: Pfizer, Bristol-Myers Squibb, UCB and Celgene, Maya H Buch Consultant of: Pfizer; AbbVie; Eli Lilly; Gilead Sciences, Inc.; Merck-Serono; Sandoz; and Sanofi, Grant/research support from: Pfizer, Roche, and UCB, Kunihiro Yamaoka Speakers bureau: AbbVie, Actelion Pharmaceuticals Japan, Asahikasei Pharma Corp, Astellas Pharma, AYUMI Pharma Co, Boehringer Ingelheim Japan, Bristol-Myers Squibb, Chugai Pharma, Daiichi Sankyo, Eisai Pharma, Eli Lilly, GlaxoSmithKline, Gilead G.K., Hisamitsu Pharma Co., Janssen Pharma, Mitsubishi-Tanabe Pharma, MSD, Nippon Kayaku, Nippon Shinyaku, Ono Pharma, Otsuka Pharma, Pfizer, Sanofi, and Takeda Industrial Pharma, Consultant of: Asahikasei Pharma Corp., AbbVie, Gilead G.K., Pfizer, Astellas Pharma Inc, Eli Lilly Japan K.K., and Japan Tobacco Inc., Grant/research support from: Takeda Industrial Pharma, Pfizer, Astellas Pharma, Daiichi Sankyo, Eli Lilly, Eisai Pharma, Teijin Pharma, MSD, Shionogi, Chugai Pharma, Nippon Kayaku, Mitsubishi-Tanabe Pharma, and AbbVie, Cianna Leatherwood Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Alena Pechonkina Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Iyabode Tiamiyu Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Deyuan Jiang Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Lei Ye Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Robin Besuyen Shareholder of: Galapagos BV, Employee of: Galapagos BV, Daniel Aletaha Speakers bureau: AbbVie, Celgene, Lilly, Merck, Novartis, Pfizer, Sanofi Genzyme, UCB, Consultant of: AbbVie, Amgen, Celgene, Lilly, Medac, Merck, Novartis, Pfizer, Roche, Sandoz, Sanofi Genzyme, Grant/research support from: AbbVie, Novartis, Roche, Kevin Winthrop Consultant of: AbbVie, Bristol-Myers Squibb, Eli Lilly and Co., Galapagos NV, Gilead Sciences, GlaxoSmithKline, Pfizer, Roche, and UCB, Grant/research support from: AbbVie, Bristol-Myers Squibb, and Pfizer

POS0660 CONCOMITANT USE OF STATINS IN FILGOTINIB-TREATED PATIENTS WITH RHEUMATOID ARTHRITIS

Background:

The Janus kinase-1 preferential inhibitor filgotinib (FIL) improved rheumatoid arthritis (RA) signs and symptoms in phase (P)3 trials.1–3 RA elevates cardiovascular disease risk; statins are used to reduce risk.

Objectives:

To assess safety of statin and filgotinib coadministration across the clinical program.

Methods:

Patients (pts) meeting 2010 ACR/EULAR RA criteria in P2 DARWIN 1–2 (D1–2; NCT01888874, NCT01894516), P3 FINCH 1–3 (F1–3; NCT02889796, NCT02873936, NCT02886728), and long-term extensions DARWIN 3 and FINCH 4 (D3, F4; NCT02065700, NCT03025308) receiving FIL 100 mg (FIL100) QD, FIL 200 mg QD (FIL200), adalimumab (ADA), methotrexate (MTX), or placebo (PBO) were included. Events related to statin use were analysed as exposed by treatment received. N and % were provided.

Week (W)12 PBO-controlled safety analysis included pts receiving FIL100, FIL200, or PBO for ≤12W (D1–2, F1–2); as-treated safety analysis included pts receiving long-term FIL100 QD (n=1647), FIL200 QD (n=2267), ADA (n=325), MTX (n=416), or PBO (n=781) (D1–3, F1–4); P3 as-randomised analysis included data up to W52 (F1–3) per assigned treatment.

Results:

In each arm, similar proportions of pts took statins at baseline (9.4%–11.9%); initiation during study was low (1.2%–6.8%). Through W12 in PBO-controlled analysis, mean creatine phosphokinase (CPK; Figure 1), alanine aminotransferase (ALT), and aspartate aminotransferase (AST) levels were similar regardless of statin use and remained within normal levels across all arms.

Mean baseline ALT and AST levels were 20–23 and 20–22 U/L, respectively; at W12, ALT and AST ranged from 22–24 and 20–25 U/L, respectively. Graded CPK, ALT, and AST elevations are in Table 1.

Table 1.

Graded laboratory abnormalities at week 12 by baseline statin use in PBO-controlled analysis

ConcomitantNoneFIL200(n=68)FIL100(n=95)PBO(n=93)FIL200 (n=709)FIL100(n=693)PBO(n=688)CPK increased*598281562549537G1 (≤2.5×ULN)10 (16.9)13 (15.9)6 (7.4)71 (12.6)47 (8.6)18 (3.4)G2 (>2.5 to 5×ULN)3 (5.1)006 (1.1)2 (0.4)3 (0.6)G3 (>5 to 10×ULN)0001 (0.2)03 (0.6)G4 (>10×ULN)0001 (0.2)2 (0.4)0AST increased**689492708692684G1 (≤3.0×ULN)9 (13.2)11 (11.7)7 (7.6)97 (13.7)79 (11.4)60 (8.8)G2 (>3.0 to 5.0×ULN)0003 (0.4)2 (0.3)3 (0.4)G3 (>5.0 to 20.0×ULN)01 (1.1)02 (0.3)00G4 (>20.0×ULN)000000ALT increased**689492708692684G1 (≤3.0×ULN)13 (19.1)14 (14.9)13 (14.1)98 (13.8)92 (13.3)72 (10.5)G2 (>3.0 to 5.0×ULN)02 (2.1)010 (1.4)5 (0.7)6 (0.9)G3 (>5.0 to 20.0×ULN)0001 (0.1)01 (0.1)G4 (>20.0×ULN)000000

Data are n (%). Grading per Common Terminology Criteria for Adverse Events v4.03

*FINCH 1–2

**DARWIN 1–2, FINCH 1–2

ALT, alanine aminotransferase; AST, aspartate aminotransferase; CPK, creatine phosphokinase; csDMARD, conventional synthetic disease-modifying antirheumatic drug; FIL200/100, filgotinib 200/100 mg + csDMARDs; Grade, G; PBO, placebo; ULN, upper limit of normal.

In the long-term as-treated analysis, 1 (0.5%)/6 (3.2%)/0/0/0 treatment-emergent adverse events (AE) of myalgia occurred in pts on statins at baseline receiving FIL200/FIL100/ADA/MTX/PBO and in 12 (0.6%)/8 (0.5%)/3 (1.0%)/2 (0.5%)/1 (0.1%) pts not on statins. Muscle spasms occurred in 2 (0.9%)/3 (1.6%)/1 (3.2%)/0/1 (1.1%) pts on statins at baseline receiving FIL200/FIL100/ADA/MTX/PBO and 21 (1.0%)/8 (0.5%)/0/3 (0.8%)/1 (0.1%) pts not on statins at baseline. One patient not on statins receiving FIL200 reported rhabdomyolysis. For all treatment arms in P3 as-randomised analysis, mean LDL and HDL increased similarly from baseline (108–110 and 56–59 mg/dL, respectively) to W52 (119–130 and 59–71 mg/dL, respectively).

Conclusion:

No increases in statin-induced AEs such as muscle or liver toxicities occurred with statins and filgotinib coadministration; results are supported by a drug-drug interaction study.4 Mean LDL and HDL increased at W52 in all treatment arms.

References:

[1]Genovese et al. JAMA. 2019;322:315–25.

[2]Westhovens et al. Ann Rheum Dis. 2021; online first.

[3]Combe et al. Ann Rheum Dis. 2021; online first.

[4]Anderson et al. EULAR 2021 abstract.

Disclosure of Interests:

Peter C. Taylor Consultant of: AbbVie, Biogen, Eli Lilly, Fresenius, Galapagos, Gilead, GlaxoSmithKline, Janssen, Nordic Pharma, Pfizer, Roche, BMS, Sanofi, Celltrion, and UCB, Grant/research support from: Celgene, Eli Lilly, Galapagos, and Gilead, Christina Charles-Schoeman Consultant of: Gilead, Pfizer, and Regeneron-Sanofi, Grant/research support from: AbbVie, Bristol-Myers Squibb and Pfizer Inc, Muhsen Alani Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Mona Trivedi Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Vanessa Castellano Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Iyabode Tiamiyu Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Deyuan Jiang Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Lei Ye Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Sander Strengholt Shareholder of: Galapagos BV, Employee of: Galapagos BV, Michael Nurmohamed Speakers bureau: AbbVie, Bristol-Myers Squibb, Eli Lilly, Roche, and Sanofi, Consultant of: AbbVie, Celgene, Celltrion, Eli Lilly, Janssen, and Sanofi, Grant/research support from: AbbVie, Bristol-Myers Squibb, Celgene, Eli Lilly, Janssen, MSD, Mundipharma, Novartis, Pfizer, Roche, and Sanofi, Gerd Rüdiger Burmester Speakers bureau: AbbVie, Eli Lilly, Pfizer, and Gilead Sciences, Inc., Consultant of: AbbVie, Eli Lilly, Pfizer, and Gilead Sciences, Inc.

POS0092 HERPES ZOSTER IN THE FILGOTINIB RHEUMATOID ARTHRITIS PROGRAM

Background:

The once daily, oral Janus kinase (JAK)-1 preferential inhibitor filgotinib (FIL) improved signs and symptoms of rheumatoid arthritis (RA) in phase (P)3 trials.1-3 Patients (pts) with RA have increased herpes zoster (HZ) reactivation risk vs the general population. JAK inhibition is associated with increased infection incidence, including HZ.4

Objectives:

To assess long-term safety of FIL across the global clinical program with respect to HZ.

Methods:

Pts meeting 2010 ACR/EULAR RA criteria in a pooled analysis of P2 DARWIN 1–2 (D1–2), P3 FINCH 1–3 (F1–3), and long-term extension studies (D3, F4) were included. Placebo (PBO)-controlled as-randomised analysis included pts receiving FIL 100 mg (FIL100), FIL 200 mg (FIL200), or PBO up to week (W)12 (D1–2, F1–2); active-controlled as-randomised analysis included pts receiving FIL100, FIL200, adalimumab (ADA), or methotrexate (MTX) up to W52 (F1, F3). Long-term as-treated analysis included pts in all 7 studies receiving FIL100, FIL200, ADA, MTX, or PBO; data after re-randomisation were included and contributed to treatment received. Exposure-adjusted incidence rates (EAIR)/100 patient-years, calculated up to the last follow-up time or day, and differences with 95% confidence intervals (CIs) were calculated from the Poisson model. Logistic regression model was used for treatment-emergent (TE) HZ risk factor analysis and odds ratio (95% CI) and P value were provided.

Results:

Table 1 shows TE HZ EAIRs in a pooled analysis. Rates of HZ were lower for FIL200 vs PBO during the 12W PBO-controlled period. At 52W, HZ rates were higher for FIL200/100 vs active control. Long-term HZ rates increased for FIL200 vs FIL100.

Table 1.

EAIR of treatment-emergent herpes zoster

NPatient-years exposureEAIR(95% CI)EAIR diff(95% CI vs PBO/active control)12W PBO-controlled FIL200777179.80.6 (0.1, 3.9)−0.56 (−2.5, 1.3) FIL100788181.61.1 (0.3, 4.4)−0.02 (−2.2, 2.2) PBO781178.41.1 (0.3, 4.5)Active-controlled, as-randomiseda FIL200475439.71.4 (0.6, 3.0)0.69 (−0.7, 2.1) FIL100480443.40.9 (0.3, 2.4)0.23 (−1.1, 1.5) ADA325297.60.7 (0.2, 2.7)Active-controlled, as-randomiseda FIL200626578.01.7 (0.9, 3.2)0.65 (−0.8, 2.2) FIL100207195.01.5 (0.5, 4.8)0.46 (−1.6, 2.5) MTX416372.21.1 (0.4, 2.9)Long-term as-treatedb FIL20022674047.71.8 (1.4, 2.3)NC FIL10016472032.91.1 (0.8, 1.7)NC

aup to W52. bdata cut for LTE FINCH 4, Sept 19, 2019; DARWIN 3, April 26 2019.

ADA, adalimumab; CI, confidence interval; EAIR, exposure-adjusted incidence rate; FIL, filgotinib; MTX, methotrexate; NC, not calculated; PBO, placebo; W week.

Figure 1 shows multivariate logistic regression model of TE risk factors.

Of 104 pts with TE HZ in long-term as-treated analysis set, 5 receiving FIL200 had history of HZ; EAIR (95% CI) was 8.7 (3.6–21.0). Of 8 pts with multiple events, 3 had events of differing severity for the same HZ episode.

EAIRs (95% CI) of TE HZ in Asia were: 3.7 (1.7–8.1) FIL200, n=197; 2.8 (1.3–6.3) FIL100, n=158; 0 ADA, n=40; 2.8 (0.4–19.6) MTX, n=43; and 3.4 (0.5–23.8) PBO, n=77 in long-term as-treated population. EAIRs (95% CI) in rest of the world were: 1.6 (1.2–2.1) FIL200, n=2070; 0.9 (0.6–1.5) FIL100, n=1489; 0.8 (0.2–3.1) ADA, n=285; 0.9 (0.3–2.9) MTX, n=373; and 0.7 (0.2–2.9) PBO, n=704 for all pts as-treated.

Most TE HZ infections were mild to moderate and non-serious; 6 were serious; 2 were recurrences. No visceral TE HZ occurred across the FIL RA program; there was 1 case each of genital, disseminated, and ophthalmic HZ. The disseminated HZ occurred in a pt with prior HZ history. Lymphopenia was not associated with HZ during the PBO-controlled W12 period.

Conclusion:

HZ was more common in both FIL groups vs ADA or MTX up to 52 weeks but comparable vs PBO during the 12-week placebo-controlled period. In multivariate analyses, prior history of HZ, Asian region, and age ≥50 years were associated with increased HZ risk.

References:

[1]Genovese et al. JAMA. 2019;322:315–25.

[2]Westhovens et al. Ann Rheum Dis. 2021; online first.

[3]Combe et al. Ann Rheum Dis. 2021; online first.

[4]Higarashi and Honda. Drugs. 2020;80:1183–201.

Disclosure of Interests:

Kevin Winthrop Consultant of: AbbVie, Bristol-Myers Squibb, Eli Lilly and Co., Galapagos NV, Gilead Sciences, GlaxoSmithKline, Pfizer, Roche, and UCB, Grant/research support from: AbbVie, Bristol-Myers Squibb, and Pfizer, Maya H Buch Speakers bureau: AbbVie; Eli Lilly and Company; Gilead Sciences, Inc.; Merck-Serono; Pfizer; Roche; Sandoz; Sanofi; and UCB, Consultant of: AbbVie; Eli Lilly and Company; Gilead Sciences, Inc.; Merck-Serono; Pfizer; Roche; Sandoz; Sanofi; and UCB, Grant/research support from: AbbVie; Eli Lilly and Company; Gilead Sciences, Inc.; Merck-Serono; Pfizer; Roche; Sandoz; Sanofi; and UCB, Jeffrey Curtis Grant/research support from: AbbVie, Amgen, BMS, Corrona, Eli Lilly, Janssen, Myriad, Pfizer, Regeneron, Roche, and UCB, Gerd Rüdiger Burmester Speakers bureau: AbbVie; Eli Lilly; Pfizer; and Gilead Sciences, Inc., Consultant of: AbbVie; Eli Lilly; Pfizer; and Gilead Sciences, Inc., Daniel Aletaha Speakers bureau: AbbVie, Amgen, Bristol-Myers Squibb, Celgene, Eli Lilly, Medac, Merck, Merck Sharp & Dohme, Novartis, Pfizer, Roche, Sandoz, Sanofi/Genzyme, and UCB, Consultant of: AbbVie, Amgen, Celgene, Eli Lilly, Janssen, Medac, Merck, Novartis, Pfizer, Roche, Sandoz, and Sanofi/Genzyme, Grant/research support from: AbbVie, Merck Sharp & Dohme, Novartis, and Roche, Koichi Amano Speakers bureau: AbbVie GK, Astellas, Chugai Pharmaceutical Co. Ltd., Eli Lilly, GlaxoSmithKline KK, Pfizer Japan, Mitsubishi-Tanabe Pharma, Grant/research support from: Asahi Kasei Pharma, Alena Pechonkina Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Iyabode Tiamiyu Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Cianna Leatherwood Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Lei Ye Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Qi Gong Shareholder of: Gilead Sciences, Inc., Employee of: Gilead Sciences, Inc., Robin Besuyen Shareholder of: Galapagos, BV, Employee of: Galapagos, BV, James Galloway Speakers bureau: Pfizer, Bristol-Myers Squibb, UCB and Celgene

[Impact of cardiovascular metabolic diseases on COVID-19: review of recent progress]

The high comorbidity between cardiovascular and metabolic diseases (CVMD) and coronavirus disease 2019 (COVID-19) and the consequent high mortality and the potential risk of cardiovascular damage have brought great challenges to the clinical diagnosis and treatment of the condition. The latest studies found that advanced age, immune function defects, inflammatory factor storms and oxidative stress damage all potentially contribute to the high comorbidity of the two. Direct virus invasion, myocardial oxygen supply and demand imbalance and vascular endothelial and coagulation dysfunction may be important mechanisms for cardiovascular injury in COVID-19 patients. In addition, the expression level of ACE2 (the cell membrane receptor of SARS-CoV-2) in various organs and the peripheral blood not only mediates the direct invasion and damage of the organs, but also participates in regulation of the balance of systematic inflammation and oxidative stress, thus affecting the susceptibility and outcomes of the patients. Herein we review the recent research progress in the comorbidity between COVID-19 and CVMD and explore the mechanisms of cardiovascular damage caused by SARS-CoV-2, thus to provide a theoretical basis for the clinical diagnosis and treatment of COVID-19 with underlying CVMD.

Using real-world data to estimate the changing trends in the prevalence and incidence of type 2 diabetes mellitus in Xiamen of China from 2014 to 2019

BACKGROUND

The prevalence of diabetes is increasing worldwide. Our study aimed to estimate the changing trends in the prevalence and incidence of diagnosed type 2 diabetes mellitus (T2DM) among Xiamen residents and the floating population using real-world data.

METHOD

We used real-world data from the System of Xiamen Citizens Health Information from 2014 to 2019 to estimate the changing trends in the prevalence and incidence of diagnosed T2DM. The System included the diagnosis of diabetes and the prescription of hypoglycemic drugs. Prevalent cases of T2DM were individuals who were diagnosed with T2DM and/or using hypoglycemic drugs. Incident cases were individuals with diagnosed T2DM and/or using hypoglycemic drugs in 2014 or 2019 who had not been diagnosed and/or did not use hypoglycemic drugs in the past.

RESULTS

In 2014 and 2019, the prevalence of T2DM in Xiamen was 4.04 and 4.84%, respectively. In 2014 and 2019, the incidence rate of T2DM in Xiamen was 14.1 per 1000 person-year and 15.0 per 1000 person-year, respectively. There was a significant increase in both the prevalence (Prevalence difference: 0.80, 95%CI 0.76-0.83%, P < 0.001) and the incidence of T2DM (Incidence difference: 0.9, 95%CI 0.7-1.1, P < 0.001). in Xiamen. The prevalence and incidence of T2DM in people aged 18-39 increased significantly (P < 0.001), while the prevalence and incidence of T2DM in people aged 40-69 reduced significantly (P < 0.001).

CONCLUSIONS

There was a significant increase in the prevalence and incidence of T2DM in Xiamen from 2014 to 2019 especially among those with younger age.

[Laparoscopic circular stapled gastrointestinal anastomosis using novel device of sealed cap access after total laparoscopic gastrectomy]

Intracorporeal classic gastrointestinal anastomosis using circular stapler in totally laparoscopic gastrectomy (TLG) for gastric cancer requires intracorporeal anvil placement and suitable access for introduction of the circular stapler to the abdominal cavity without gas leak. The novel techniques for anvil placement have been updated, but there is no progress for proper access for circular stapler. In the study, intracorporeal circular-stapled gastrointestinal anastomosis were successfully accomplished using a novel device of sealed cap access with a central hole (WLB-60/70-60/100, Wuhan Widerep Medical Instrument Co.,Ltd, China) customized to the incision protection retractor for the simple and accessible introduction of the circular stapler and anvil under the optimal maintenance of pneumoperitoneum pressure in TLG. In these 3 cases, there was no gas leakage and the pneumoperitoneum was well maintained when performing the gastrointestinal anastomosis, and there was no transition to laparotomy or other anastomosis techniques. The result suggests that the sealed cap access could be a novel choice for introduction of the circular stapler to the abdominal cavity in order to obtain laparoscopic circular-stapled gastroin-testinal anastomosis in TLG.

Gene Module Analysis Reveals Cell-Type Specificity and Potential Target Genes in Autism's Pathogenesis

Multiple genetic factors contribute to the pathogenesis of autism spectrum disorder (ASD), a kind of neurodevelopmental disorder. Genes were usually studied separately for their associations with ASD. However, genes associated with ASD do not act alone but interact with each other in a network module. The identification of these modules is the basis for the systematic understanding of the pathogenesis of ASD. Moreover, ASD is characterized by highly pathogenic heterogeneity, and gene modules associated with ASD are cell-type-specific. In this study, based on the single-nucleus RNA sequencing data of 41 post-mortem tissue samples from the prefrontal cortex and anterior cingulate cortex of 19 ASD patients and 16 control individuals, we applied sparse module activity factorization, a matrix decomposition method consistent with the multi-factor and heterogeneous characteristics of ASD pathogenesis, to identify cell-type-specific gene modules. Then, statistical procedures were performed to detect highly reproducible cell-type-specific ASD-associated gene modules. Through the enrichment analysis of cell markers, 31 cell-type-specific gene modules related to ASD were further screened out. These 31 gene modules are all enriched with curated ASD risk genes. Finally, we utilized the expression patterns of these cell-type-specific ASD-associated gene modules to build predictive models for ASD. The excellent predictive performance also proved the associations between these gene modules and ASD. Our study confirmed the multifactorial and cell-type-specific characteristics of ASD pathogeneses. The results showed that excitatory neurons such as L2/3, L4, and L5/6-CC play essential roles in ASD's pathogenic processes. We identified the potential ASD target genes that act together in cell-type-specific modules, such as NRG3, KCNIP4, BAI3, PTPRD, LRRTM4, and LINGO2 in the L2/3 gene modules. Our study offers new potential genomic targets for ASD and provides a novel method to study gene modules involved in the pathogenesis of ASD.

Impact of maternal infection with hepatitis B virus on pregnancy complications and neonatal outcomes for women undergoing assisted reproductive technology treatment: A population-based study

The aim of this study was to investigate the impact of maternal hepatitis B virus (HBV) status on pregnancy complications and neonatal outcomes for women undergoing assisted reproductive technology (ART). A total of 7,011 pregnancies achieved by ART were included from a population-based database involving 523,111 pregnancies. Exposures of hepatitis B surface antigen (HBsAg) and hepatitis B e antigen (HBeAg) among pregnant women were routinely tested at the first antenatal visit for all pregnancies. We collected pregnancy complications (e.g., gestational diabetes mellitus [GDM], intrahepatic cholestasis of pregnancy [ICP]), neonatal outcomes and confounding variables from the same database. Univariate and multivariate analyses by adjusting confounders were conducted to evaluate the impact of maternal HBV infection. Prevalence of HBsAg seropositivity (HBsAg+) was 11.34% (95% CI 10.6-12.1) and that of HBsAg and HBeAg co-seropositivity (HBsAg+HBeAg+) was 2.55% (2.1-3.0) among included population. Compared with HBsAg-group, ICP risk in the HBsAg+group was higher (4.03% vs. 1.79%; adjusted odds ratio [aOR] 2.49, 1.65-3.77). Similarly, ICP prevalence in the HBsAg+HBeAg+ group was higher than that in the HBsAg-HBeAg- group (6.47% vs. 1.61%; aOR 4.78, 2.28-9.98). No associations were found between maternal HBV infection (i.e., HBsAg+, HBsAg+HBeAg+, or HBsAg+HBeAg-) and other adverse outcomes for women undergoing ART (i.e., GDM, pre-eclampsia, placental previa, premature separation of placenta, premature rupture of membranes, preterm birth and low birthweight) in this study. In conclusion, maternal HBV infection (HBsAg+or HBsAg+HBeAg+) probably increase ICP risk, but may not associate with other pregnancy complications or neonatal outcomes for pregnant women who underwent ART.

Peri-prostatic adipose tissue measurements using MRI predict prostate cancer aggressiveness in men undergoing radical prostatectomy

OBJECTIVES

To evaluate the effect of peri-prostatic adipose tissue (PPAT) measurements using preoperative MRI on the prediction of prostate cancer (PCa) aggressiveness in men undergoing radical prostatectomy (RP).

METHODS

We performed a retrospective study on 179 consecutive patients receiving RP from June 2016 to October 2018. Clinical characteristics were collected. PPAT measurements including peri-prostatic fat area (PPFA) and peri-prostatic fat area to prostate area (PA) ratio (PPFA/PA) were calculated by MRI. Multivariable logistic regression analysis was performed to identify independent predictors of PCa lymph node metastasis (LNM). The predictive performance was estimated through ROC curves. Nomograms were created based on the predictors.

RESULTS

Pathologic Gleason score positively correlated with digital rectal examination (DRE), PSA, PPFA/PA, P504S, and Ki-67 (all P < 0.05). ROC curves revealed that high PPFA and high PPFA/PA were associated with LNM (both P < 0.05). Multivariate analysis revealed that high PPFA/PA, pathologic Gleason score, pT stage, and Ki-67 were independently predictive of LNM. The nomograms were created and the C-index was 0.945.

CONCLUSIONS

PPFA/PA is an independent predictor for LNM along with Gleason score, pT stage, and Ki-67. PPFA/PA may help predict LNM in men undergoing RP, thus providing adjunctive information for therapeutic strategy and prognosis.

Wnt antagonist secreted frizzled-related protein I (sFRP1) may be involved in the osteogenic differentiation of periodontal ligament cells in chronic apical periodontitis

AIM

To explore the mechanism of secreted frizzled-related protein I (sFRP1) involvement in the osteogenic differentiation of human periodontal ligament cells (hPDLCs) under inflammatory conditions.

METHODOLOGY

hPDLCs were cultured in an osteogenic differentiation-inducing medium (odi) and subjected to the stimulation of Porphyromonas gingivalis lipopolysaccharide (P. gingivalis LPS) with or without the inhibition of sFRP1. Quantitative real-time polymerase chain reaction, Western blot and enzyme-linked immunosorbent assay were carried out to evaluate the expression of osteogenic markers as well as the classic Wnt signalling pathway. Periapical periodontitis was induced in Wistar rats to further confirm the effect of sFRP1 inhibitor on bone loss in vivo. After the Shapiro-Wilk normality test, data were analysed by Student's paired t-test or one-way Anova test with a P value less than 0.05 as the level of statistical significance.

RESULTS

Significantly decreased mRNA and protein expression of osteogenic markers were detected in hPDLCs treated with P. gingivalis LPS during osteogenic induction (P < 0.001). Increased expression of sFRP1 was observed (P < 0.01), whilst Wnt/β-catenin signalling pathway was inhibited by the addition of P. gingivalis LPS (P < 0.01). After the addition of the sFRP1 inhibitor, the decrease of osteogenic markers (P < 0.05) and the inhibition of Wnt/β-catenin signalling pathway (P < 0.05) were reversed significantly. The animal experiment further confirmed that the sFRP1 inhibitor significantly reduced bone loss of periapical lesions in vivo (P < 0.0001).

CONCLUSIONS

Wnt antagonist sFRP1 was involved in the osteogenic differentiation of hPDLCs under inflammation. Modulation of the Wnt/β-catenin signalling pathway through the inhibition of sFRP1 may offer a new perspective on the treatment of chronic apical periodontitis.

[Epidemic situation of malaria in Lishui City from 2013 to 2018]

OBJECTIVE

To analyze the epidemic situation and epidemiological characteristics of malaria in Lishui City from 2013 to 2018, so as to provide the evidence for formulating the malaria control strategy.

METHODS

The data pertaining malaria cases in Lishui City from 2013 to 2018 were captured from National Notifiable Communicable Disease Reporting System and the Information System for Parasitic Diseases Control and Prevention, and the epidemiological features of malaria cases were analyzed.

RESULTS

A total of 119 malaria cases were reported in Lishui City from 2013 to 2018, including 101 cases with falciparum malaria (84.87%), 6 cases with vivax malaria (5.04%), 8 cases with ovale malaria (6.72%), and 4 cases with mixed infection (3.36%). Among the 119 cases, there were one local case with blood transfusion-induced malaria and 118 cases with over- seas imported malaria. There were 98.32% of the imported malaria cases acquiring infection in African countries, and most cases were reported in Qingtian County (60.50%) and Liandu District (22.69%). In addition, 86.55% of the malaria cases were detected in individuals at ages of 20 to 50 years, and most cases were found in oversea workers (52.94%) and businessmen (38.65%).

CONCLUSIONS

Most of the malaria cases in Lishui City are imported from Africa, and the monitoring and health education pertaining to malaria control knowledge requires to be intensified among high-risk populations.

Plasma concentration of ZCCHC14 contributes to prognostic efficacy in intracerebral hemorrhage patients

OBJECTIVE

Intracerebral hemorrhage (ICH) is a devastating cerebrovascular disease of the central nervous system. In this study, we aimed to investigate plasma ZCCHC14 expression and its correlation with the diagnosis, prognosis, and clinical features of ICH.

PATIENTS AND METHODS

The plasma levels of ZCCHC14, TNF-α, IL-6, and IL-10 were dynamically detected among 130 ICH patients and 60 corresponding healthy subjects. A receiver operating characteristic (ROC) curve was used for the statistical analysis of the diagnostic and prognostic efficacy of ICH.

RESULTS

Hypertension (p = 0.005), ARB application (p = 0.014), ZCCHC14 level (p < 0.001), TNF-α (p < 0.001), IL-6 (p < 0.001), and IL-10 (p < 0.001) were found to be different between ICH patients and healthy controls. Multiple logistic regression analysis indicated that ZCCHC14 levels at admission were significantly different between the two groups (p = 0.002, OR = 0.440, 95% CI 0.290-0.718). Plasma ZCCHC14 levels dynamically changed, increasing at admission, and peaking on day 7. ZCCHC14 was a potential diagnostic marker for ICH (AUC = 0.953, p < 0.0001, 95% CI: 0.901-1.004) with a specificity and sensitivity of 84.6% and 95.5%, respectively. The plasma ZCCHC14 level was negatively correlated with IL-6 concentration (p = 0.024, r = -0.311) but was positively correlated with IL-10 concentration (p = 0.041, r = 0.298). Furthermore, the plasma ZCCHC14 level was correlated with hypertension (p = 0.005), GOS (p = 0.025), bleeding volume (p < 0.001), midline shift (p = 0.003), and poor outcome (p = 0.006). The low ZCCHC14 expression group had poorer outcomes (death or severe disability) than high ZCCHC14 expression group (Breslow, p < 0.001).

CONCLUSIONS

We found that the plasma ZCCHC14 level might be a potential biomarker for both the early diagnosis of and prediction of outcomes in ICH.

Prognosis of COVID-19 in patients with vein thrombosis: a systematic review and meta-analysis

OBJECTIVE

SARS-CoV-2 is currently affecting millions of humans worldwide, thus contributing to the COVID-19 pandemic. Thromboembolic events have a higher incidence among patients with COVID-19, but there are few reports on the relationship between the prognosis of COVID-19 patients and thromboembolic events. The objectives of this meta-analysis were to explore the relationship between the prognosis of COVID-19 patients and thromboembolic events.

MATERIALS AND METHODS

Medline (PubMed), the Web of Science, Embase, and the Cochrane Library were searched for case-control studies that included data on vein thrombosis in patients with COVID-19 and were published in English, between January 1 and July 25, 2020. According to the inclusion and exclusion criteria, the included data were confirmed, the prognoses of patients with and without concurrent thromboembolic events were compared, and the odds ratio (OR) was used as the effect size.

RESULTS

Eighteen studies (2,030 patients) were included. Thromboembolic events complicated a total of 609 COVID-19 patients. The combined OR of the mortality of COVID-19 patients with thromboembolic events was 1.93 (95% CI: 1.13-3.27), that of ICU treatment rate was 2.63 (95% CI: 1.49-4.67), and that of treatment with invasive mechanical ventilation was 3.14 (95% CI: 1.97-5.02).

CONCLUSIONS

As compared with COVID-19 patients with and without thromboembolism, the mortality, ICU treatment rate, and invasive mechanical ventilation treatment rate of COVID-19 patients with thromboembolism were found to be increased significantly, and the prognosis was worse.

Arrhythmia in patients with severe coronavirus disease (COVID-19): a meta-analysis

OBJECTIVE

Many studies have reported arrhythmia to be associated with coronavirus disease (COVID-19), but no meta-analysis has explored whether arrhythmia is related to COVID-19 severity. Therefore, the purpose of this study was to evaluate arrhythmia in patients with severe and non-severe COVID-19 during the current COVID-19 pandemic.

MATERIALS AND METHODS

We searched PubMed, Embase, Web of Science, and the Cochrane Library for case control studies that were published between January 1 and July 25, 2020, and that had data on arrhythmia in patients with COVID-19. Random effects model was used with the odds ratio as the effect size. The frequency of arrhythmia was compared between COVID-19 patients with and without the composite endpoint of severity. We also determined the pooled prevalence of arrhythmia in patients with COVID-19. Publication bias and heterogeneity were considered by using subgroup analyses, meta-regression, and the trim and fill method.

RESULTS

A total of 1553 patients with COVID-19 were included in the 5 articles we obtained. Of these, 349 cases (22.47%) and 1204 cases (77.53%) were severely ill and non-severely ill inpatients with COVID-19 pneumonia, respectively. There were 790 (50.87%) male patients. A total of 105 cases (30.09%) of severely ill inpatients with COVID-19 pneumonia had arrhythmia complications, and 34 cases (2.82%) of non-severely ill inpatients with COVID-19 pneumonia had arrhythmia complications. We found arrhythmia to be significantly associated with severely ill inpatients with COVID-19 pneumonia, with a pooled odds ratio of 17.97 (95% CI (11.30, 28.55), p<0.00001).

CONCLUSIONS

This study showed that the incidence of arrhythmia in patients with severe COVID-19 was greater than that of those with non-severe COVID-19. Patients with severe COVID-19 had a higher risk of arrhythmia complications, which further showed that COVID-19 may be a risk factor for arrhythmia and that the incidence of arrhythmia may increase with the progression of the disease. More importantly, this meta-analysis graded the reliability of evidence for further basic and clinical research into arrhythmia in patients with COVID-19.

Effect of Deep Margin Elevation on CAD/CAM-Fabricated Ceramic Inlays

CLINICAL RELEVANCE

Using the deep margin elevation technique in preparations extending beyond the cemento-enamel junction appears to be beneficial in maintaining structural integrity of CAD/CAM-fabricated feldspathic ceramic inlays.

SUMMARY

Objective: To evaluate the effect of deep margin elevation on structural and marginal integrity of ceramic inlays.Methods and Materials: Forty extracted human third molars were collected and randomly separated into four groups (n=10/group). In group 1 (enamel margin group), the gingival margin was placed 1 mm supragingival to the cemento-enamel junction (CEJ). In group 2 (cementum margin group), the gingival margin was placed 2 mm below the CEJ. In group 3 (glass ionomer [GI] margin group), the gingival margin was placed 2 mm below the CEJ, and then the margin elevated with GI to the CEJ. In group 4 (resin-modified glass ionomer [RMGI] margin group), the gingival margin was placed 2 mm below the CEJ, and then the margin elevated with RMGI to the CEJ. Standardized ceramic class II inlays were fabricated with computer-aided design/computer-aided manufacturing and bonded to all teeth, and ceramic proximal box heights were measured. All teeth were subjected to 10,000 cycles of thermocycling (5°C/55°C) and then underwent 1,200,000 cycles of vertical chewing simulation at 50 N of force. Ceramic restorations and marginal integrity were assessed with a Hirox digital microscope. The Fisher exact test (two-tailed) with adjusted p-values (α=0.05) and logistic regression were used for statistical analysis.Results: The cementum margin group had a significantly higher ceramic fracture rate (90%) compared to other groups (10% in enamel margin and GI margin groups, p=0.007; 0% in RMGI group, p<0.001). Logistic regression showed that with increased ceramic proximal box heights, the probability of ceramic fracture increased dramatically.Conclusion: Deep marginal elevation resulted in decreased ceramic fracture when preparation margins were located below the CEJ. There was no difference found between margin elevation with GI or RMGI. Increased heights of ceramic proximal box may lead to an increased probability of ceramic fracture.

Prediction of chronic kidney disease progression used by calcineurin inhibitor concentration and estimated glomerular filtration rate early after liver transplantation

Background

Chronic kidney disease (CKD) is a common late complication in liver-transplanted patients who have received long-term therapy with calcineurin inhibitors (CNIs). Aims: To analyze kidney disease progression after liver transplantation.

Methods

We analysed the clinical data of adult single-organ liver transplant recipients performed at our centre between October 2003 and September 2009. The patients with the estimated glomerular filtration rate (eGFR) greater than 60 ml/min/1.73 m² before surgery were included in the study.

Results

69 patients with complete follow-up data were analysed. We found that eGFR at 1 or 2 years after liver transplantation correlated well with eGFR at 5 years. In addition, our results showed that patients whose eGFR declined below 60 at 2 years after liver transplantation would develop an irreversible renal injury in the following years. At 2 years, 12 patients had an eGFR less than 60, which were maintained in 11 patients at 5 years (Sensitivity = 11/12, 91.67%; Specificity = 57/58, 98.28%, Youden's index = 89.95%). The annual rate of eGFR reduction of the tacrolimus group was greater than that of the tacrolimus sparing group based on the value-time variation curve in our study. Moreover, the tacrolimus concentration influenced the CKD progression at 1 and 2 years with an under the ROC curve of 0.73 and 0.78 when Youden's index was at its maximum and the tacrolimus concentrations were 8.55 and 5.96 ng/ml, respectively.

Conclusion

We confirmed that eGFR at 2 years after liver transplantation is useful for observing a meaningful change in eGFR and renal damage. Obtaining the appropriate serum concentration of an early decrease of the dose of CNIs and transforming non-nephrotoxic immunosuppressants would help improve renal function to prevent CKD progression and end-stage renal disease (ESRD).

scHinter: imputing dropout events for single-cell RNA-seq data with limited sample size

MOTIVATION

Single-cell RNA-sequencing (scRNA-seq) is fast and becoming a powerful technique for studying dynamic gene regulation at unprecedented resolution. However, scRNA-seq data suffer from problems of extremely high dropout rate and cell-to-cell variability, demanding new methods to recover gene expression loss. Despite the availability of various dropout imputation approaches for scRNA-seq, most studies focus on data with a medium or large number of cells, while few studies have explicitly investigated the differential performance across different sample sizes or the applicability of the approach on small or imbalanced data. It is imperative to develop new imputation approaches with higher generalizability for data with various sample sizes.

RESULTS

We proposed a method called scHinter for imputing dropout events for scRNA-seq with special emphasis on data with limited sample size. scHinter incorporates a voting-based ensemble distance and leverages the synthetic minority oversampling technique for random interpolation. A hierarchical framework is also embedded in scHinter to increase the reliability of the imputation for small samples. We demonstrated the ability of scHinter to recover gene expression measurements across a wide spectrum of scRNA-seq datasets with varied sample sizes. We comprehensively examined the impact of sample size and cluster number on imputation. Comprehensive evaluation of scHinter across diverse scRNA-seq datasets with imbalanced or limited sample size showed that scHinter achieved higher and more robust performance than competing approaches, including MAGIC, scImpute, SAVER and netSmooth.

AVAILABILITY AND IMPLEMENTATION

Freely available for download at https://github.com/BMILAB/scHinter.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

The Impact of Prior Season Vaccination on Subsequent Influenza Vaccine Effectiveness to Prevent Influenza-related Hospitalizations Over 4 Influenza Seasons in Canada

BACKGROUND

Recent studies have demonstrated the possibility of negative associations between prior influenza vaccines and subsequent influenza vaccine effectiveness (VE), depending on season and strain. We investigated this association over 4 consecutive influenza seasons (2011-2012 through 2014-2015) in Canada.

METHODS

Using a matched test-negative design, laboratory-confirmed influenza cases and matched test-negative controls admitted to hospitals were enrolled. Patients were stratified into 4 groups according to influenza vaccine history (not vaccinated current and prior season [referent], vaccinated prior season only, vaccinated current season only, and vaccinated both current and prior season). Conditional logistic regression was used to estimate VE; prior vaccine impact was assessed each season for overall effect and effect stratified by age (<65 years, ≥65 years) and type/subtype (A/H1N1, A/H3N2, influenza B).

RESULTS

Overall, mainly nonsignificant associations were observed. Trends of nonsignificant decreased VE among patients repeatedly vaccinated in both prior and current season relative to the current season only were observed in the A/H3N2-dominant seasons of 2012-2013 and 2014-2015. Conversely, in 2011-2012, during which B viruses circulated, and in 2013-2014, when A/H1N1 circulated, being vaccinated in both seasons tended to result in a high VE in the current season against the dominant circulating subtype.

CONCLUSIONS

Prior vaccine impact on subsequent VE among Canadian inpatients was mainly nonsignificant. Even in circumstances where we observed a trend of negative impact, being repeatedly vaccinated was still more effective than not receiving the current season's vaccine. These findings favor continuation of annual influenza vaccination recommendations, particularly in older adults.

CLINICAL TRIALS REGISTRATION

NCT01517191.