IDentification of Factors Predictive of Tofacitinib's Survival in Patient With Rheumatoid Arthritis

The duration of tofacitinib drug survival was calculated as: date of permanent drug discontinuation minus date of first taking the drug + 1. If a participant did not present with the event of interest, that is they did not have a permanent drug discontinuation record during the study, then they were censored at the time of their last visit. Kaplan-Meier method was used for estimation.

PCS is a 13 item questionnaire around thoughts and emotions experienced during pain, with each question scored on a 5-point scale ranging from 0 (not at all) to 4 (all the time), where higher scores indicated worse condition. There are three subscales of PCS: rumination (thinking deeply) [4 items], exaggeration [3 items] and helplessness [6 items]. Total scores and domain scores are derived by summing the scores of the individual items. Overall possible Total PCS score range is 0 to 52, where higher scores signifies worse condition. A total score of greater than or equal to (>=) 20 indicates a clinically relevant level of catastrophising. Mean change from baseline in total PCS score is reported in this outcome measure.

CSQ evaluates cognitive coping, has 21 items and 5 domains in its French version. For each item, participants rate the frequency of their use of each coping strategy on a 4- point Likert-type scale (0= Never, 1= sometimes, 2= often and 3= very often). Domain scores are derived by summing the individual items scores that make-up the domain. Five domains with score range: distraction (5 items, score range: 0 to 15), catastrophising (4 items, score range: 0 to 12), distancing from pain (4 items, score range: 0 to 12), ignoring pain sensations (5 items, score range: 0 to 15) and praying (3 items, score range: 0 to 9). For each domain, higher scores indicated worse condition. Mean change from baseline in CSQ domain scores are reported in this outcome measure.

DAS28-4 ESR is a composite endpoint, calculated using 4 variables (represented by '4' in the name). Components includes: Tender Joint Count (TJC) with 28 joints assessed, Swollen Joint Count (SJC) with 28 joints assessed, ESR (millimeter per hours [mm/h]) and Patient Global Assessment (PtGA) recorded on 100 mm VAS (scores ranging 0 [no disease activity] to 100 mm [maximum disease activity], higher scores=more disease activity). The calculation of DAS28-4 ESR is as follows: 0.56 * square root (sqrt) (TJC) + 0.28 * sqrt (SJC) + 0.70* In (ESR) + 0.014* (PtGA); where ln = natural logarithm. Total DAS28-4 score range: 0 to 9.4, higher score=more disease activity. DAS28(ESR) score of <=3.2 indicates LDA. Percentage of participants with DAS28-4 ESR <=3.2 is presented in this outcome measure.

DAS28-4 CRP is a composite endpoint, calculated using 4 variables (represented by '-4' in the name). Components includes: TJC with 28 joints assessed, SJC with 28 joints assessed, CRP (mg/L) and PtGA recorded on 100 mm VAS (scores ranging 0 [no disease activity] to 100 mm [maximum disease activity], higher scores=more disease activity). The calculation of DAS28-4 CRP is as follows: 0.56 * sqrt (TJC) + 0.28 * sqrt (SJC) + 0.36 * ln (CRP+l) + 0.014*PtGA + 0.96; where ln = natural logarithm. Total DAS28-4 score range: 0 to 9.4, higher score=more disease activity. DAS28(CRP) score of <=3.2 indicates LDA. Percentage of participants with DAS28-4 CRP <=3.2 is presented in this outcome measure.

SDAI is a composite end point, calculated as TJC + SJC (both using 28 joints) + PtGA (0-10 cm scale, higher score=more disease activity) + Physician Global Assessment (PhGA) (0-10 cm scale, higher score=more disease activity) + CRP (milligrams per deciliter [mg/dL]). The SDAI score is classified into four categories: remission (<=3.3), low (>3.3-11), moderate (>11-26), and high (>26) disease activity. The SDAI score ranges from 0 to 86 where higher scores indicates high disease activity. SDAI score of <=11 indicates LDA. Percentage of participants with SDAI <=11 is presented in this outcome measure.

CDAI is a composite end point, calculated as TJC + SJC (both using 28 joints) + PtGA (0-10 cm scale, higher score=more disease activity) + PhGA (0-10 cm scale, higher score=more disease activity). The CDAI score is classified into four categories: remission (<=2.8), low (>2.8-10), moderate (>10-22), and high (>22) disease activity. The CDAI score ranges from 0 to 76, where higher scores indicates high disease activity. CDAI score of <=10 indicates LDA. Percentage of participants with CDAI <=10 is presented in this outcome measure.

DAS28-4 ESR is a composite endpoint, calculated using 4 variables (represented by '-4' in the name). Components includes: TJC with 28 joints assessed, SJC with 28 joints assessed, ESR (mm/h) and PtGA recorded on 100 mm VAS (scores ranging 0 [no disease activity] to 100 mm [maximum disease activity], higher scores=more disease activity). The calculation of DAS28-4 ESR is as follows: 0.56 * sqrt (TJC) + 0.28 * sqrt (SJC) + 0.70* In(ESR) + 0.014* (PtGA); where ln = natural logarithm. Total DAS28-4 score range: 0 to 9.4, higher score=more disease activity. DAS28(ESR) score of <2.6 indicates LDA. Percentage of participants with DAS28-4 ESR <2.6 is presented in this outcome measure.

DAS28-4 CRP is a composite endpoint, calculated using 4 variables (represented by '-4' in the name). Components includes: TJC with 28 joints assessed, SJC with 28 joints assessed, CRP (mg/L) and PtGA recorded on 100 mm VAS (scores ranging 0 [no disease activity] to 100 mm [maximum disease activity], higher scores=more disease activity). The calculation of DAS28-4 CRP is as follows: 0.56 * sqrt (TJC) + 0.28 * sqrt (SJC) + 0.36 * ln (CRP+l) + 0.014*PtGA + 0.96; where ln = natural logarithm. Total DAS28-4 score range: 0 to 9.4, higher score=more disease activity. DAS28(CRP) score of <2.6 indicates LDA. Percentage of participants with DAS28-4 CRP <2.6 is presented in this outcome measure.

SDAI is a composite end point, calculated as TJC + SJC (both using 28 joints) + PtGA (0-10 cm scale, higher score=more disease activity) + PhGA (0-10 cm scale, higher score=more disease activity) + CRP (mg/dL). The SDAI score is classified into four categories: remission (<=3.3), low (>3.3-11), moderate (>11-26), and high (>26) disease activity. The SDAI score ranges from 0 to 86 where higher scores indicates high disease activity. SDAI score of <=3.3 indicates LDA. Percentage of participants with SDAI <=3.3 is presented in this outcome measure.

CDAI is a composite end point, calculated as TJC + SJC (both using 28 joints) + PtGA (0-10 cm scale) + PhGA (0-10 cm scale). The CDAI score is classified into four categories: remission (<=2.8), low (>2.8-10), moderate (>10-22), and high (>22) disease activity. The CDAI score ranges from 0 to 76, where higher scores indicates high disease activity. CDAI score of <=2.8 indicates LDA. Percentage of participants with CDAI <=2.8 is presented in this outcome measure.

ACREULAR 2011 Boolean criteria is derived as: ACR Remission = 1, if SJC (using 28 joints) <=1, TJC (using 28 joints) <=1, CRP <=1 mg/dL, and PtGA (0 to 10 cm scale, higher score=more disease activity) <=1 centimeter (cm) otherwise ACR remission =0. Higher scores indicated greater affection due to disease activity. Percentage of participants with remission according to ACR-EULAR 2011 Boolean criteria is reported in this outcome measure.

DAS28-4 ESR is a composite endpoint, calculated using 4 variables (represented by '-4' in the name). Components includes: TJC with 28 joints assessed, SJC with 28 joints assessed, ESR (mm/h) and PtGA recorded on 100 mm VAS (scores ranging 0 [no disease activity] to 100 mm [maximum disease activity], higher scores=more disease activity). The calculation of DAS28-4 ESR is as follows: 0.56 * sqrt (TJC) + 0.28 * sqrt (SJC) + 0.70* In(ESR) + 0.014* (PtGA); where ln = natural logarithm. Total DAS28-4 score range: 0 to 9.4, higher score=more disease activity. Mean change from baseline in DAS28-4 ESR score at months 3, 6, 12, 18 and 24 is presented in this outcome measure.

DAS28-4 CRP is a composite endpoint, calculated using 4 variables (represented by '-4' in the name). Components includes: TJC with 28 joints assessed, SJC with 28 joints assessed, CRP (mg/L) and PtGA recorded on 100 mm VAS (scores ranging 0 [no disease activity] to 100 mm [maximum disease activity], higher scores=more disease activity). The calculation of DAS28-4 CRP is as follows: 0.56 * sqrt (TJC) + 0.28 * sqrt (SJC) + 0.36 * ln (CRP+l) + 0.014*PtGA + 0.96; where ln = natural logarithm. Total DAS28-4 score range: 0 to 9.4, higher score=more disease activity. Mean change from baseline in DAS28-4 CRP score at months 3, 6, 12, 18 and 24 is presented in this outcome measure.

28 EULAR criteria to evaluate participant's response to treatment is categorised as good responder, moderate responder and non-responder based on participants DAS28-4 CRP score and decrease in DAS28-4 CRP compared to initial pre-treatment visit (Visit 0 [V0]). Good response: DAS28 at treatment visit <= 3.2 and decrease in DAS 28 compared to V0 >1.2. Moderate responder: DAS28 at treatment visit > 3.2 to <= 5.1 or > 5.1 and decrease in DAS 28 compared to V0 >1.2 DAS28 at treatment visit <= 3.2 or > 3.2 to <= 5.1 and decrease in DAS 28 compared to V0 >0.6 and <=1.2. Non-responder: DAS28 at treatment visit > 5.1 and decrease in DAS 28 compared to V0 >0.6 to <=1.2 DAS28 at treatment visit <= 3.2 or > 3.2 to <= 5.1 or > 5.1 and decrease in DAS 28 compared to V0 <=0.6. Number of participants responding to treatment by considering DAS28 (EULAR criterion) is reported in this outcome measure.

TJC (28) is the count of the total number of tender joints out of a possible 28 joints which are checked at the visit. TJC was used to measure the pain and inflammation in the joints. Mean change from baseline in TJC is reported in this outcome measure.

SJC (28) is the count of the total number of swollen joints out of a possible 28 joints which are checked at the visit. SJC was used to measure the pain and inflammation in the joints. Mean change from baseline in TJC is reported in this outcome measure.

FiRST is used to detect fibromyalgia in participants with diffuse rheumatic pain. It is a questionnaire which contains 6 items: diffuse pain, painful symptoms, fatigue, sleep and cognitive disorders, nonpainful abnormal sensations and functional somatic symptoms. Each of the 6 items contain yes or no response, and a positive answer to 5 out of the 6 answers can detect fibromyalgia. At each specified time points number of participants with "Yes" or "No" to fibromyalgia is reported in this outcome measure.

Morning stiffness is a common symptom of rheumatoid arthritis. The duration of morning stiffness is measured in minutes. Mean change from baseline in morning stiffness is reported in this outcome measure.

Girerd questionnaire is used to evaluate participant's treatment compliance and is made up of 6 yes or no questions. lf the participant responds "no" to all questions, it is considered as participant is a good complier. lf the participant responds "yes" once or twice, it is considered as participant is a minor complier. lf the participant responds "yes" three times or more, it is considered as participant is a no-complier.

EQ-5D-3L is a standardized, participant administered measure of self-reported instrument used to evaluate health-related quality of life. It consists of two parts: EQ-5D index score (Part I) and the EQ-VAS (Part II). For Part I, i.e. EQ-5D-3L index score, participants rated their current health state on 5 single-item dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression with each dimension having three levels of function:. 1=no problems, 2=some problems and 3=extreme problems. Scoring formula developed by EuroQol Group assigns a utility value for each domain in the profile. Score was transformed and results in a total index score range of 0 to 1.00; higher scores indicating a better health condition. Mean change from baseline in EQ-5D-3L index score is reported in this outcome measure.

SF-12 was a participant reported outcome survey that represented overall health status by measuring 8 health-related aspects of an individual: Physical Functioning (PF), Role-Physical (RP), Bodily Pain (BP), General Health (GH), Vitality (VT), Social Functioning (SF), Role-Emotional (RE) and Mental Health (MH). The score range for each of the 8 health aspects was from 0 (poor health) to 100 (better health), higher scores indicating good health condition. Responses on the SF-12 were also used to calculate 2 summary scores: Physical component score (PCS) and mental component score (MCS). The score range for each of these 2 summary scores was from 0 (poor health) to 100 (better health), where 100 indicated good health condition. Mean change from baseline in PCS and MCS scores are reported in this outcome measure.

FACIT-Fatigue questionnaire consists of 13 self-evaluated questions. Each question has a response of values 0 (not at all) to 4 (very much). Score for each question is calculated by first reversing negatively stated-items (subtracting the response from '4' except for the 2 positively stated-items) and then summing the raw (0-4) scores. The FACIT-Fatigue total score is derived by summing the response to each question which gives a value between 0 (worse score) and 52 (best score). Higher scores represent better participant's status (less fatigue). Mean change from baseline in FACIT-Fatigue total score is reported in this outcome measure.

Number of participants who discontinued tofacitinib due to tolerance issue.