Relapse (Yes/No) was defined as meeting both the eosinophil histology relapse criterion and the dysphagia symptom relapse criterion. Eosinophil histology relapse was defined as an eosinophil count of greater than or equal to(> or =) 15 per high-power field (eos/hpf) from at least 2 of 3 levels of the esophagus. Dysphagia symptom relapse was defined as having at least 4 days of dysphagia (with answer 'Yes' for question 2 in DSQ [Dysphagia Symptom Questionnaire]) in the 2-week period prior to the scheduled visit, as determined by the DSQ.

SHP621-301 (NCT02605837) BOS Full Responder FAS consisted of all randomized participants who were determined to be Full responders to the BOS treatment during the induction study SHP621-301 (NCT02605837) and who received at least 1 dose of IP in the SHP621-302 (NCT02736409) study.

Long-term response from SHP621-301 baseline defined: histologic response,defined as peak eosinophil count of less than or equal to(< or =)6/HPF across all available esophageal levels at Week 36 and Dysphagia symptom response 1,defined as >or= 30 percent(%) reduction in DSQ combined score(Questions [Q] 2+3) from baseline of SHP621-301 to Week 36.DSQcontain 4 questions,all participants used diary,responded to Q1(didyoueatsolid food),Q2(did food pass slowly or get stuck). If participant's answer to Q2 was No,diary ended for day. If participant answered Yes,he/she advanced to Q3(didyouhavetodoanything to make food go downorget relief), Q4(extent to which participant experienced pain while swallowing).DSQ combined score= ([sum of points from Q2+3 in daily DSQ]×14)/Number of diaries reported with non-missing data.Scale range was0-2 forQ2 and 0-4 forQ 3.Scale range for DSQ combined score was 0-84.Highervaluesrepresentingworseoutcome.Negative change from baseline indicates symptoms decreased.

SHP621-301 (NCT02605837) BOS Non responder FAS included all randomized participants who did not respond or partially responded to assigned BOS treatment in the induction study SHP621-301 (NCT02605837) and who received at least 1 dose of IP in the SHP621-302 (NCT02736409) study.

Long-term response from baseline defined: histologic response, defined as peak eosinophil count of less than or equal to(< or =) 6/HPF across all available esophageal levels at Week 36 and Dysphagia symptom response 1,defined as >or= 30 percent(%) reduction in DSQ combined score (Questions [Q] 2+3) from baseline of SHP621-301 to Week 36.DSQcontain 4 questions,all participants used diary,responded to Q1(did you eat solid food),Q2(did food pass slowly or get stuck). If participant's answer to Q2 was No,diary ended for day. If participant answered Yes,he/she advanced to Q3(didyouhavetodoanything to make food go downorget relief), Q4(extent to which participant experienced pain while swallowing).DSQ combined score= ([sum of points from Q2+3 in daily DSQ]×14)/Number of diaries reported with non-missing data.Scale range was0-2 forQ2 and 0-4 forQ 3.Scale range for DSQ combined score was 0-84. Higher values representing worse outcome.Negative change from baseline indicates symptoms decreased.

SHP621-301 (NCT02605837) BOS Non responder FAS included all randomized participants who did not respond or partially responded to assigned BOS treatment in the induction study SHP621-301 (NCT02605837) and who received at least 1 dose of IP in the SHP621-302 (NCT02736409) study.

Histology response was defined as a peak eosinophil count of \

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study.

Dysphagia symptom response with respect to the baseline of induction study (SHP621-301 [NCT02605837]) was defined as 30% reduction in DSQ combined score (questions 2+3) at Week 12 and Week 36 of current study from baseline of the induction study. DSQ contain 4 questions, all participants used a diary and responded to Q1 (did you eat solid food), Q2 (did food pass slowly or get stuck). If participant's answer to Q2 was No, then diary ended for that day. If participant answered Yes,he/she advanced to Q3 (did you have to do anything to make food go down or get relief) and Q4 (extent to which participant experienced pain while swallowing). DSQ combined score= ([sum of points from Q2+3 in daily DSQ] × 14)/Number of diaries reported with non-missing data. Scale range was 0-2 for Q2 and 0-4 for Q3. Scale range for DSQ combined score was 0-84. Higher values representing worse outcome. Negative change from baseline indicates symptoms decreased.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study.

Dysphagia symptom response with respect to the baseline of current study (SHP621-302 [NCT02736409]) was defined as 30% reduction in DSQ combined score (questions 2+3) at Week 12 and Week 36 from baseline of the current study. DSQ contain 4 questions, all participants used diary and responded to Q1 (did you eat solid food), Q2 (did food pass slowly or get stuck). If participant's answer to Q2 was "No", then diary ended for that day. If participant answered "Yes",he/she advanced to Q3 (did you have to do anything to make food go down or get relief) and Q4 (extent to which participant experienced pain while swallowing). DSQ combined score= ([sum of points from Q2+3 in daily DSQ] × 14)/Number of diaries reported with non-missing data. Scale range was 0-2 for Q2 and 0-4 for Q 3. Scale range for DSQ combined score was 0-84. Higher values representing a worse outcome. A Negative change from baseline indicates symptoms decreased.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study.

Endoscopic findings with separate evaluations of the proximal and distal esophagus were recorded with respect to 5 categories: 1) exudates or plaques (grade 0-2); 2) fixed esophageal rings (grade 0-3); 3) edema (grade 0-2); 4) furrows (grade 0-2); and 5) strictures (grade 0-1). An endoscopy score for each category was calculated and summed for each anatomic location (proximal and distal). The minimum and maximum endoscopy score was 0 and 10 points respectively for each location (proximal and distal) and the total endoscopy score was the sum of the scores for the proximal and distal locations (maximum total score of 20 points respectively). The higher score indicated worse appearance. A negative change from baseline indicates that appearance improved.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study. Here the number of participants analyzed signifies participants who had an evaluable Induction Study (SHP621-301 [NCT02605837]) baseline and assessment at specific category/time point.

Endoscopic findings with separate evaluations of the proximal and distal esophagus were recorded with respect to 5 categories: 1) exudates or plaques (grade 0-2); 2) fixed esophageal rings (grade 0-3); 3) edema (grade 0-2); 4) furrows (grade 0-2); and 5) strictures (grade 0-1). An endoscopy score for each category was calculated and summed for each anatomic location (proximal and distal). The minimum and maximum endoscopy score was 0 and 10 points respectively for each location (proximal and distal) and the total endoscopy score was the sum of the scores for the proximal and distal locations (maximum total score of 20 points respectively). The higher score indicated worse appearance. A negative change from baseline indicates that appearance improved.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study. Here the number of participants analyzed signifies participants who had an evaluable current Study (SHP621-302 [NCT02736409]) baseline and assessment at specific category/time point.

Change from induction study (SHP621-301 [NCT02605837]) baseline in peak eosinophil count at Week 12 and Week 36 of current study was reported.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study. Here the number of participants analyzed signifies participants who had an evaluable Induction Study (SHP621-301 [NCT02605837]) baseline and assessment at specific category/time point.

Change from current study (SHP621-302 [NCT02736409]) baseline in peak eosinophil count at Week 12 and Week 36 was reported.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study. Here the number of participants analyzed signifies participants who had an evaluable current Study (SHP621-302 [NCT02736409]) baseline and assessment at specific category/time point.

Proportion of participants who had Peak Eosinophil Count less than (<) 15/HPF at Week 12 and Week 36 were reported.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study. Here the number of participants analyzed signifies participants who were evaluable for this outcome measure at specific category.

Proportion of participants who had Peak Eosinophil Count less than or equal to (< or =) 1/High-Powered Field (HPF) at Week 12 and Week 36 were reported.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study.

Change from induction study (SHP621-301 [NCT02605837]) baseline in peak eosinophil count for each available esophageal level (proximal, mid, distal) at Week 12 and Week 36 of current study were reported.

FAS included all randomized participants who received at least one dose of a SHP621-302 (NCT02736409) investigational product. Here the number of participants analyzed signifies participants who had an evaluable Induction Study (SHP621-301 [NCT02605837]) baseline and assessment at specific category/time point.

Change from current study (SHP621-302 [NCT02736409]) baseline in peak eosinophil count for each available esophageal level (proximal, mid, distal) at Week 12 and Week 36 were reported.

FAS included all randomized participants who received at least one dose of a SHP621-302 (NCT02736409) investigational product. Here the number of participants analyzed signifies participants who had an evaluable current Study (SHP621-302 [NCT02736409]) baseline and assessment at specific category/time point.

Eight histopathologic epithelial features (basal layer hyperplasia, eosinophil peak, abscesses, surface layering, dilated intercellular spaces, surface alteration, dyskeratotic epithelial cells, lamina propria fibrosis) are scored on a 4-point scale (0=normal, 3=worst) for both the severity of the abnormality (i.e., grade) and the amount of tissue affected by the abnormality (i.e. stage). Thus each of the 3 levels had a minimum score of 0 and maximum possible score of 24, and a possible total grade or stage score of 72 for a maximum combined score of 144. Combined total score ratio (TSR) =(proximal TSR + mid TSR + distal TSR)/N, where N is the number of non missing sections for TSR. A negative change from baseline indicates that epithelial inflammation decreased.

FAS included all randomized participants who received at least one dose of a SHP621-302 (NCT02736409) investigational product. Here the number of participants analyzed signifies participants who had an evaluable Induction Study (SHP621-301 [NCT02605837]) baseline and assessment at specific category/time point.

Eight histopathologic epithelial features (basal layer hyperplasia, eosinophil peak, abscesses, surface layering, dilated intercellular spaces, surface alteration, dyskeratotic epithelial cells, lamina propria fibrosis) are scored on a 4-point scale (0=normal, 3=worst) for both the severity of the abnormality (i.e., grade) and the amount of tissue affected by the abnormality (i.e. stage). Thus each of the 3 levels had a minimum score of 0 and maximum possible score of 24, and a possible total grade or stage score of 72 for a maximum combined score of 144. Combined total score ratio (TSR) =(proximal TSR + mid TSR + distal TSR)/N, where N is the number of non missing sections for TSR. A negative change from baseline indicates that epithelial inflammation decreased.

FAS included all randomized participants who received at least one dose of a SHP621-302 (NCT02736409) investigational product. Here the number of participants analyzed signifies participants who had an evaluable current Study (SHP621-302 [NCT02736409]) baseline and assessment at specific category/time point.

Dysphagia symptom response from the baseline of SHP621-301, was defined as >or=50% reduction in DSQ combined score (questions [Q] 2+3) from baseline of the induction study at Week 12 and 36 of current study. DSQ contained 4Q, all participants used diary, and responded to Q1(did you eat solid food) and Q2(didfood passslowly or get stuck). If participant's answer to Q2 was 'No', then diary ended for that day. If a participant answered 'Yes', he/she advanced to Q3(did you have to do anything to make food go down or get relief) and 4(extent to which participant experienced pain while swallowing). DSQ combined score= ([sum of points from Q2+3 in daily DSQ]×14)/ Number of diaries reported with non-missing data. Scale range was 0-2 for Q2 and 0-4 for Q3, with higher values representing a worse outcome. Scale range for DSQ combined score was 0-84, with higher values representing a worse outcome. Anegative change from baseline indicates that symptoms decreased.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study.

Dysphagia symptom response from the baseline of SHP621-302, was defined as >or=50% reduction in DSQ combined score (questions 2+3) from baseline of the induction study at Week 36 of current study. DSQ contained 4Q, all participants used diary, and responded to Q1 (did you eat solid food) and Q2 (did food pass slowly or get stuck). If participant's answer to Q2 was 'No', then diary ended for that day. If a participant answered 'Yes', he/she advanced to Q3 (did you have to do anything to make food go down or get relief) and 4(extent to which participant experienced pain while swallowing). DSQ combined score= ([sum of points from Q2+3 in daily DSQ]×14)/ Number of diaries reported with non-missing data. Scale range was 0-2 for Q2 and 0-4 for Q3, with higher values representing a worse outcome. Scale range for DSQ combined score was 0-84, with higher values representing a worse outcome. A negative change from baseline indicates that symptoms decreased.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study.

DSQ contain 4 questions (Q), all participants used a diary, responded to Q1 (did you eat solid food), Q2 (did food pass slowly or get stuck). If participant's answer to Q2 was 'No', then diary ended for that day. If participant answered 'Yes', he/she advanced to Q3 (did you have to do anything to make food go down or get relief), Q4 (extent to which participant experienced pain while swallowing). DSQ combined score= ([sum of points from Q2+3 in daily DSQ] × 14)/Number of diaries reported with non-missing data. Scale range was 0-2 for Q2 and 0-4 for Q3. Scale range for DSQ combined score was 0-84. Higher values representing a worse outcome. A Negative change from baseline indicates symptoms decreased.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study. Here the number of participants analyzed signifies participants who were evaluable for this outcome measure.

DSQ contain 4 questions, all participants used a diary, responded to Q1 (did you eat solid food), Q2 (did food pass slowly or get stuck). If participant's answer to Q2 was 'No', then diary ended for that day. If participant answered 'Yes', he/she advanced to Q3 (did you have to do anything to make food go down or get relief), Q4 (extent to which participant experienced pain while swallowing). DSQ combined score= ([sum of points from Q2+3 in daily DSQ] × 14)/Number of diaries reported with non-missing data. Scale range was 0-2 for Q2 and 0-4 for Q3. Scale range for DSQ combined score was 0-84. Higher values representing a worse outcome. A Negative change from baseline indicates symptoms decreased. Change in DSQ combined score (Questions 2+3) from baseline of current study at Week 36 was reported.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study. Here the number of participants analyzed signifies participants who were evaluable for this outcome measure.

Percent change in DSQ combined score from Induction Study (SHP621-301 [NCT02605837]) baseline to current study (SHP621-302 [NCT02736409]) final treatment period (Week 36) was reported.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study. Here the number of participants analyzed signifies participants who were evaluable for this outcome measure.

Percent change in DSQ combined score from baseline to final treatment period (Week 36) of current study was reported.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study. Here the number of participants analyzed signifies participants whose baseline DSQ combined score was not zero. Percent change cannot be calculated if baseline DSQ combined score is zero.

Overall binary response I was defined as if participant had peak eosinophil count of <=6/HPF across all esophagus levels and achieved a minimum of 30% reduction in DSQ combined score from baseline of the induction study (SHP621-301 [NCT02605837]) at Week 12 and Week 36. DSQ combined score= ([sum of points from questions 2+3 in the daily DSQ]×14)/ Number of diaries reported with non-missing data. Scale range was 0 - 2 for question 2 and 0 - 4 for question 3, with higher values representing a worse outcome. Scale range for DSQ combined score was 0 - 84, with higher values representing a worse outcome. A negative change from baseline indicates that symptoms decreased.

FAS included all randomized participants who received at least one dose of a SHP621-302 (NCT02736409) investigational product.

Overall binary response I was defined as if participant had peak eosinophil count of <=6/HPF across all esophagus levels and achieved a minimum of 30% reduction in DSQ combined score from baseline at Week 12 and Week 36. DSQ combined score= ([sum of points from questions 2+3 in the daily DSQ]×14)/ Number of diaries reported with non-missing data. Scale range was 0 - 2 for question 2 and 0 - 4 for question 3, with higher values representing a worse outcome. Scale range for DSQ combined score was 0 - 84, with higher values representing a worse outcome. A negative change from baseline indicates that symptoms decreased.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study.

Overall binary response II was defined as if participant had peak eosinophil count of <=6/HPF across all esophagus levels and achieved a minimum of 50% reduction in DSQ combined score from baseline of the SHP621-301 study at Week 12 and Week 36. DSQ combined score= ([sum of points from questions 2+3 in the daily DSQ]×14)/ Number of diaries reported with non-missing data. Scale range was 0 - 2 for question 2 and 0 - 4 for question 3, with higher values representing a worse outcome. Scale range for DSQ combined score was 0 - 84, with higher values representing a worse outcome. A negative change from baseline indicates that symptoms decreased.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study.

Overall binary response II was defined as if participant had peak eosinophil count of <=6/HPF across all esophagus levels and achieved a minimum of 30% reduction in DSQ combined score from baseline at Week 12 and Week 36. DSQ combined score= ([sum of points from questions 2+3 in the daily DSQ]×14)/ Number of diaries reported with non-missing data. Scale range was 0 - 2 for question 2 and 0 - 4 for question 3, with higher values representing a worse outcome. Scale range for DSQ combined score was 0 - 84, with higher values representing a worse outcome. A negative change from baseline indicates that symptoms decreased.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study.

DSQ contained 4 questions, all participants used a diary, and responded to Questions 1 (did you eat solid food) and 2 (did food pass slowly or get stuck). If the participant's answer to Question 2 was 'No', the diary ended for that day. If a participant answered 'Yes', he/she advanced to Questions 3 (did you have to do anything to make the food go down or get relief) and 4 (extent to which the participant experienced pain while swallowing). DSQ + pain score was calculated by summing the scores of responses to questions 2, 3, and 4 by using following formula: DSQ + pain score= ([sum of points from questions 2+3+4 in the daily DSQ] ×14)/ Number of diaries reported with non-missing data. Scale range was 0 - 2 for question 2, 0 - 4 for question 3 and 0 - 4 for question 4, with higher values representing a worse outcome. Scale range for DSQ + pain score was 0 - 140, with higher values representing a worse outcome. A negative change from baseline indicates that symptoms decreased.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study. Here the number of participants analyzed signifies participants who were evaluable for this outcome measure.

DSQ contained 4 questions, all participants used a diary, and responded to Questions (Q) 1 (did you eat solid food) and 2 (did food pass slowly or get stuck). If the participant's answer to Q2 was 'No', the diary ended for that day. If a participant answered 'Yes', he/she advanced to Q3 (did you have to do anything to make the food go down or get relief) and 4 (extent to which the participant experienced pain while swallowing). DSQ + pain score was calculated by summing the scores of responses to Questions 2, 3, and 4. Question 1 was excluded from the DSQ + pain score. DSQ + pain score=(Sum of points from questions 2+3+4 in the daily DSQ)*14 Days/Number of dairies reported with non-missing data. Scale range was 0 - 2 for Q2, 0 - 4 for Q3 and 0 - 4 for Q4, with higher values representing a worse outcome. Scale range for DSQ + pain score was 0 - 140, with higher values representing a worse outcome. A negative change from baseline indicates that symptoms decreased.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study. Here the number of participants analyzed signifies participants who were evaluable for this outcome measure.

DSQ contained 4 questions, all participants used a diary, and responded to Questions 1 (did you eat solid food) and 2 (did food pass slowly or get stuck). If the participant's answer to Question 2 was 'No', the diary ended for that day. If a participant answered 'Yes', he/she advanced to Questions 3 (did you have to do anything to make the food go down or get relief) and 4 (extent to which the participant experienced pain while swallowing). DSQ pain score was calculated by summing the scores of responses to Question 4 only. DSQ pain score=(sum of points from questions 4 in the daily DSQ)*14 days/Number of diaries reported with non-missing data. Scale range was 0 - 4 for question 4, with higher values representing a worse outcome. Scale range for DSQ pain score was 0 - 56, with higher values representing a worse outcome. A negative change from baseline indicates that symptoms decreased.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study. Here the number of participants analyzed signifies participants who were evaluable for this outcome measure.

DSQ contained 4 questions, all participants used a diary, and responded to Questions 1 (did you eat solid food) and 2 (did food pass slowly or get stuck). If the participant's answer to Question 2 was 'No', the diary ended for that day. If a participant answered 'Yes', he/she advanced to Questions 3 (did you have to do anything to make the food go down or get relief) and 4 (extent to which the participant experienced pain while swallowing). DSQ pain score was calculated by summing the scores of responses to Question 4 only. DSQ pain score=(sum of points from questions 4 in the daily DSQ)*14 days/Number of diaries reported with non-missing data. Scale range was 0 - 4 for question 4, with higher values representing a worse outcome. Scale range for DSQ pain score was 0 - 56, with higher values representing a worse outcome. A negative change from baseline indicates that symptoms decreased.

FAS consisted of all randomized participants who received at least 1 dose of investigational product in the SHP621-302 (NCT02736409) study. Here the number of participants analyzed signifies participants who were evaluable for this outcome measure.

An AE was any untoward medical occurrence in a clinical investigation participant administered a pharmaceutical product and that does not necessarily have a causal relationship with this treatment. TEAEs were defined as AEs that started or deteriorated on or after the date of the first dose of double-blind IP in SHP621-302 and through the safety follow-up contact, or 31 days after the last dose of IP for participants who did not have a safety follow-up contact.

Safety Set consisted of all participants who received at least 1 dose of investigational product.

The sites for DXA measurement were the lumber spine (L1-L4 preferred) and total body less head. DXA scans for bone mineral density (BMD) and body composition measurements were done only for adolescent participants aged 11-17 years based on the age group in the SHP621-301 study and were completed throughout the SHP621-302 study even if participants turned 18 years. Baseline of SHP621-301 is determined at Week 0 in the SHP621-301 study. Here in this outcome measure DXA measurement of L1-L4 were reported. Z-score indicates the number of standard deviations away from a reference population in the same age range and with the same sex. A Z-score of 0 is equal to the mean. Negative numbers indicate values lower than the mean and positive numbers indicate values higher than the mean.

Safety set consisted of all participants who received at least 1 dose of investigational product. Here the number of participants analyzed signifies participants who were evaluable for this outcome measure.

The sites for DXA measurement were the lumber spine (L1-L4 preferred) and total body less head. DXA scans for bone mineral density (BMD) and body composition measurements were done only for adolescent participants aged 11-17 years based on the age group in the SHP621-301 study and were completed throughout the SHP621-302 study even if participants turned 18 years. Here in this outcome measure DXA measurement of L1-L4 were reported. Z-score indicates the number of standard deviations away from a reference population in the same age range and with the same sex. A Z-score of 0 is equal to the mean. Negative numbers indicate values lower than the mean and positive numbers indicate values higher than the mean.

Safety set consisted of all participants who received at least 1 dose of investigational product. Here the number of participants analyzed signifies participants who were evaluable for this outcome measure.

The sites for DXA measurement were the lumber spine (L1-L4 preferred) and total body less head. DXA scans for bone mineral density (BMD) and body composition measurements were done only for adolescent participants aged 11-17 years based on the age group in the SHP621-301 study and were completed throughout the SHP621-302 study even if participants turned 18 years. Baseline of SHP621-301 is determined at Week 0 in the SHP621-301 study. Here in this outcome measure DXA measurement of whole body (except head) were reported. Z-score indicates the number of standard deviations away from a reference population in the same age range and with the same sex. A Z-score of 0 is equal to the mean. Negative numbers indicate values lower than the mean and positive numbers indicate values higher than the mean.

Safety set consisted of all participants who received at least 1 dose of investigational product. Here the number of participants analyzed signifies participants who were evaluable for this outcome measure.

The sites for DXA measurement were the lumber spine (L1-L4 preferred) and total body less head. DXA scans for bone mineral density (BMD) and body composition measurements were done only for adolescent participants aged 11-17 years based on the age group in the SHP621-301 study and were completed throughout the SHP621-302 study even if participants turned 18 years. Here in this outcome measure DXA measurement of whole body (except head) were reported. Z-score indicates the number of standard deviations away from a reference population in the same age range and with the same sex. A Z-score of 0 is equal to the mean. Negative numbers indicate values lower than the mean and positive numbers indicate values higher than the mean.

Safety set consisted of all participants who received at least 1 dose of investigational product. Here the number of participants analyzed signifies participants who were evaluable for this outcome measure.

Eosinophil histology relapse was defined as an eosinophil count of >or=15 eos/HPF from at least 2 of 3 levels of the esophagus.

SHP621-301 (NCT02605837) BOS Full Responder FAS consisted of all randomized participants who were determined to be Full responders to the BOS treatment during the induction study SHP621-301 (NCT02605837) and who received at least 1 dose of IP in the SHP621-302 (NCT02736409) study.