1. A randomised crossover study to evaluate patient preference for Rituximab SC or IV MO28457 (PrefMab)
Stuart Osborne, Roche PDMA Statistics BBS BES Joint Meeting 22-Nov-2018

2. Overview Background information Study results Regulatory submission
Study design
Primary objective
Approach to analysis & sample size
Development of the Patient Preference Questionnaire (PPQ) for PrefMab
PPQ
PPQ analysis population definitions in Protocol & Statistical Analysis Plan
Study results
Treatment exposure
Preference & Strength of preference at Cycle 8
Reasons for preference at Cycle 8
Regulatory submission
PrefMab included in Rituximab SC submission to FDA
Subsequent FDA interaction
Final wording in the US label
References

3. Cross-over Study Design
BACK
Cross-over Study Design
8x R-chemo (CHOP/CVP/Benda)
Rituxan SC
(1400 mg)
RITUXAN IV
(375 mg/m2)
CTSQ and RASQ R A N D O M I Z E
1:1
2 years follow-up
Untreated
FL or
DLBCL (N=743)
Arm A (n=372)
Arm B (n=371)
PPQ
end of cycle
Primary endpoint
Secondary endpoints
Patient Preference for IV or SC using Questionnaire (PPQ)
Administration time
CTSQ
RASQ
Safety
Efficacy – CR, EFS, DFS, PFS, OS

4. Primary objective
To evaluate the proportion of patients indicating an overall preference via the Patient Preference Questionnaire (PPQ) for either the SC or the IV route of rituximab administration

5. Approach to analysis and sample size
Descriptive analysis only - the binary primary endpoint (patient preference for rituximab SC) will be estimated with a two-sided 95% confidence interval.
Sample size can be estimated so that the lower limit of the 95% confidence interval is higher than 50%.
For PrefMab, it was anticipated (no reference data) that the proportion of patients preferring rituximab SC will be approximately 60%.
The study planned a high sample size (actual was less but still relatively high). Protocol stated that with the planned sample size the 95% CI for SC preference rate of 60% = (56.4%, 63.6%).
Software used: nQuery Version 7

6. Development of the Patient Preference Questionnaire (PPQ) for PrefMab
A specific PPQ for PrefMab was developed based on the patient preference questionnaire from the PrefHer trial (HER2-positive early breast cancer trial with a cross-over design assessing patient preference for SC vs IV dosing) (Pivot et al., 2013).
The PrefHer measure was developed through standardized iterative methodology:
Clinician input generated topics that might affect preference, which were formulated into questions
Questions modified to ensure relevance and intelligibility
Questions tested with patient volunteers who’d previously received treatment for breast cancer
Further modifications to improve clarity and remove ambiguities
The result was a questionnaire that assessed patient preference using questions that were relevant and easy to comprehend. This was subsequently applied to the PrefMab study.
© 2017, Genentech / Proprietary information – Please do not copy, distribute or use without prior written consent.

7. PrefMab PPQ (Patient Preference Questionnaire)
You have now received the drug Rituximab in two different ways:
through a thin plastic tube and a needle that was put directly into a vein in your arm, called an intravenous or IV infusion.
given through a needle injected into your abdomen (or stomach) area, called a subcutaneous or SC injection.
Please answer the following questions about your experiences and your preferences. There are not any right or wrong answers.
1) All things considered which method of administration did you prefer?
IV SC No preference
2) If you have a preference for one of the administration routes, how strong is this preference?
Very strong Fairly strong Not very strong
3) If you have a preference for one of the administration routes, what are the TWO main reasons for your preference?
Feels less emotionally distressing
Requires less time in the clinic
Lower level of injection-site pain
Feels more comfortable during administration
Other reason; please specify: ……………………..………………………
© 2017, Genentech / Proprietary information – Please do not copy, distribute or use without prior written consent.

8. PrefMab PPQ population definitions
Primary population: ITT
743 patients (372 in Arm A and 371 in Arm B), was defined as all patients randomized into the study.
Within the ITT population, patients were allocated to treatment groups according to their original randomization schedule.
Efficacy and patient-related outcome (patient preference, RASQ, and CTSQ) endpoints were summarized based on the ITT population.
Sensitivity analyses: modified ITT (mITT)
645 patients (323 in Arm A and 322 in Arm B),
required the patient to have received both routes of administration and to have completed the primary question in the PPQ at either Cycle 6 or Cycle 8.
A patient was considered to have received both* routes of administration if that patient has received at least one administration of each route during cycles 2 through 8.
*cycle 1, always IV, did not count for the both routes of administration condition
© 2017, Genentech / Proprietary information – Please do not copy, distribute or use without prior written consent.

9. Treatment exposure (Safety population)
Cycles completed
Arm A (N=371)
N (%)
Arm B (N=369)
Total (N=740)
Cycle 1
371 (100%)
369 (100%)
740 (100%) …
Cycle 4
349 (94.1%)
349 (94.6%)
698 (94.3%)
Cycle 8
311 (83.8%)
309 (83.7%)
620 (83.8%)

10. Preference & strength of preference (mITT) at Cycle 8
Variable
Arm A (n = 323)
Arm B (n = 322)
Total (N = 645)
Patients completing the Patient Preference Questionnaire at Cycle 8, n
293
298
591
Patients who prefer SC, n (%)
226 (77.1%)
251 (84.2%)
477 (80.7%)
95% confidence interval
77.9, 81.8
79.6, 88.2
77.3, 83.8
Strength of preference, n (%)
Very strong
117 (39.9%)
114 (38.3%)
231 (39.1%)
Fairly strong
92 (31.4%)
112 (37.6%)
204 (34.5%)
Not very strong
16 (5.5%)
25 (8.4%)
41 (6.9%)
Patient didn't answer the question
1 (0.3%)
1 (0.2%)
Patients who prefer IV, n (%)
37 (12.6%)
29 (9.7%)
66 (11.2%)
7 (2.4%)
15 (5.0%)
22 (3.7%)
15 (5.1%)
7 (2.3%)
6 (2.0%)
21 (3.6%)
Patients with no preference, n (%)
30 (10.2%)
18 (6.0%)
48 (8.1%)

11. Reasons for preference (mITT) at Cycle 8
Variable
Arm A (n = 323)
Arm B (n = 322)
Total (N = 645)
Patients completing the Patient Preference Questionnaire at Cycle 8, n
293
298
591
Patients who prefer SC, n (%)
226 (77.1%)
251 (84.2%)
477 (80.7%)
Feels less emotionally distressing
81 (27.6%)
91 (30.5%)
172 (29.1%)
Requires less time in the clinic
205 (70.0%)
205 (68.8%)
410 (69.4%)
Lower level of injection site pain
40 (13.7%)
52 (17.4%)
92 (15.6%)
Feels more comfortable during administration
98 (33.4%)
119 (39.9%)
217 (36.7%)
Other reason
17 (5.8%)
19 (6.4%)
36 (6.1%)
Patient didn't answer the question
5 (1.7%)
10 (1.7%)
Patients who prefer IV, n (%)
37 (12.6%)
29 (9.7%)
66 (11.2%)
16 (5.5%)
15 (5.0%)
31 (5.2%)
3 (1.0%)
4 (1.3%)
7 (1.2%)
18 (6.1%)
14 (4.7%)
32 (5.4%)
25 (8.5%)
44 (7.4%)
7 (2.3%)
12 (2.0%)
1 (0.3%)
1 (0.2%)
Patients with no preference, n (%)
30 (10.2%)
18 (6.0%)
48 (8.1%)

12. PrefMab included in Rituximab SC submission to FDA
FDA requested a face-to-face meeting to discuss Patient Preference results (11Jan2017)
FDA request:
Discuss the patient preference results (across all four studies) in reference to Rituxan SC.
Present the methods, results, and analysis (including data completeness) from the patient preference questionnaires.
Roche presentation based on requirements in FDA Guidance on Patient Preference Measures in Device Labeling.
© 2017, Genentech / Proprietary information – Please do not copy, distribute or use without prior written consent.

13. Subsequent FDA interaction
FDA followed up meeting with an Information Request on the PrefMab PPQ and PrefHer PPQ requesting:
Qualitative summary reports, if available, and any other literature supporting their content validity.
Quantitative summary reports, if available, and any other literature supporting other measurement properties (reliability, validity, ability to detect change, etc).
Training materials and administration instructions (for both sites and patients), as well as user manuals, if available.
Documentation of any measures that were taken to ensure adequate comprehension, i.e. tutorials or quizzes provided to patients prior to or during the trial.
Documentation of translation techniques.
Analysis results by country and indication (FL, DLBCL).
© 2018, Genentech / Proprietary information – Please do not copy, distribute or use without prior written consent.

14. Final wording in the US label (section 14 Clinical Studies)
14.4 Patient Experience Previously untreated adult patients outside of the United States with CD20+ diffuse large B-cell lymphoma (DLBCL) or CD20+ follicular non-Hodgkin’s lymphoma (FL) Grades 1, 2, or 3a were randomized to receive a standard chemotherapy regimen (CHOP, CVP, or bendamustine) and either RITUXAN HYCELA 1,400mg/23,400 Units at Cycles 2–4 (after the first cycle with intravenous rituximab) or a rituximab product by intravenous infusion at Cycles 1–4. After the fourth cycle, patients were crossed over to the alternative route of administration for the remaining 4 cycles. After Cycle 8, 477 of 620 patients (77%) reported preferring subcutaneous administration of RITUXAN HYCELA over intravenous rituximab and the most common reason was that administration required less time in the clinic. After Cycle 8, 66 of 620 patients (11%) preferred rituximab intravenous administration and the most common reason was that it felt more comfortable during administration. Forty eight of 620 patients (7.7%) had no preference for the route of administration. Twenty nine subjects of 620 (4.7%) received Cycle 8 but did not complete the preference questionnaire.

15. References
“Preference for subcutaneous or intravenous administration of rituximab among patients with untreated CD20+ diffuse large B-cell lymphoma or follicular lymphoma: results from a prospective, randomized, open-label, crossover study (PrefMab)” / Rummel et al / Annals of Oncology 2017; 28:
“Validation of a treatment satisfaction questionnaire in non-Hodgkin lymphoma: assessing the change from intravenous to subcutaneous administration of rituximab” / Theodore-Oklota et al / Patient Preference & Adherence 2016; 10:
“Preference for subcutaneous or intravenous administration of trastuzumab in patients with HER2-positive early breast cancer (PrefHer): an open-label randomized study” / Pivot et al / Lancet Oncology 2013; 14:
“Patients preferences for SC trastuzumab versus conventional IV infusion for the adjuvant treatment of HER2-positive EBC: final analysis of 488 patients in the international, randomized, two-cohort PrefHer study” / Pivot et al / Annals of Oncology 2014; 25:
“Implications of subcutaneous or intravenous delivery of trastuzumab; further insight from patient interviews in the PrefHer study” / Fallowfield et al / The Breast 2015; 24:
“Switching between intravenous (IV) and subcutaneous (SC) trastuzumab: Safety results from the PrefHer trial” / Gligorov et al / The Breast 2017; 34: 89-95
“Efficacy and safety of subcutaneous trastuzumab and intravenous trastuzumab as part of adjuvant therapy for HER2-positive early breast cancer: Final analysis of the randomised, two-cohort PrefHer study” / Pivot et al / European Journal of Cancer 2017; 86: 82-90