Real-world data of immune tolerance induction using recombinant factor VIII Fc fusion protein in patients with severe haemophilia A with inhibitors at high risk for immune tolerance induction failure: A follow-up retrospective analysis.

Dear Editor,

Prophylactic factor VIII (FVIII) replacement is the current standard of care for severe haemophilia A but approximately 25%–40% of patients develop inhibitors against exogenous FVIII, rendering FVIII replacement therapy ineffective. Eradication of high‐titre inhibitors involves immune tolerance induction (ITI): repeated, long‐term administration of high‐dose FVIII.

Recombinant FVIII Fc fusion protein (rFVIIIFc [ELOCTATE®, Sanofi, Waltham, MA]) is the first extended half‐life FVIII approved for haemophilia A. Case reports and an initial retrospective chart review suggest that rFVIIIFc ITI may lead to faster tolerization than ITI with standard FVIII concentrates. , This letter reports final clinical outcomes of 29 patients (19 included in the initial analysis) with severe haemophilia A undergoing ITI with rFVIIIFc in a real‐world setting.

We performed a retrospective review of patient charts at 13 sites across the United States and Canada, using previously published methods. Briefly, de‐identified clinical data were collected from patients with severe haemophilia A and historical high‐titre inhibitors, who began first‐time or rescue ITI with rFVIIIFc between July 2014 and February 2018 and had ≥3 months of exposure to rFVIIIFc ITI. Rescue ITI patients were defined as patients who had failed at least one previous ITI attempt. Tolerization was defined as a negative Bethesda titre (<0.6 BU/mL), normal FVIII recovery (≥66% of expected) and rFVIIIFc half‐life ≥6 hours.

Altogether, 29 rFVIIIFc ITI patients were identified: 10 first‐time (Table 1) and 19 rescue patients (Table 2). Median (range) age at initiation of rFVIIIFc ITI was 1.4 (0.4–4.3) years for first‐time and 6.5 (1.6–48.9) years for rescue patients. Of the 10 first‐time ITI patients, 3 had peak inhibitor titres >200 BU/mL (accepted risk factor for ITI failure), while 8 had inhibitor titres >10 BU/mL at ITI start (traditionally considered a risk factor for ITI failure, although many clinicians are disputing this). All rescue ITI patients were considered high risk for ITI failure; all had previously undergone ITI, 9 had peak inhibitor titres >200 BU/mL and 16 had an inhibitor for >2 years.

Table 1

First‐time ITI patients ,

Patient	FVIII genotype	Inhibitor titre (BU/mL)		Factor brand being used when inhibitor developed	rFVIIIFc ITI regimen	Time (weeks)							Current titre (BU/mL)	Current status
		Historical peak (pre‐ITI)	Immediately pre‐rFVIIIFc ITI			Weekly factor usage (IU/kg)	Inhibitor diagnosis to start of rFVIIIFc ITI	From start of ITI to				Duration of rFVIIIFc ITI
								Negative Bethesda titre §	Normal recovery ¶	Half‐life ≥6 h ‡‡	Tolerization ‡‡
1‐9 §§	Intron−22	38.4	20.8	rFVIIIFc (Eloctate, Sanofi, Waltham, MA)	200 IU/kg q.d.	1400	6	3	NR	3	3	3	Negative	rFVIIIFc prophylaxis
1‐1	Missense	51.7	51.7	rFVIIIFc (Eloctate, Sanofi, Waltham, MA)	85 IU/kg q.d.	595	11	4	10	21	21	21	Negative	rFVIIIFc prophylaxis
1‐8 §§	Intron−22	25.6	25.6	rFVIIIFc (Eloctate, Sanofi, Waltham, MA)	200 IU/kg q.d.	1400	18	9	NR	21	21	23	Negative	rFVIIIFc prophylaxis
1‐2	Frameshift	150.9	106.9	pdFVIII (Alphanate, Grifols Biologicals LLC, Los Angeles, CA)	110 IU/kg q.d.	770	12	24	NR	29	29	30	Negative	rFVIIIFc prophylaxis
1‐5	Intron−22	376.0	32.0	rFVIII (Advate, Baxalta US Inc, Lexington, MA)	100 IU/kg q.d.	700	41	30	56	NR	30	64	Negative	rFVIIIFc prophylaxis
1‐3	Unknown	1126.0	1126.0	rFVIII (Advate, Baxalta US Inc, Lexington, MA)	200 IU/kg q.d.	1400	1	31	NR	40	40	40	Negative	rFVIIIFc prophylaxis
1‐7 ¶¶ , †††	Intron−22	3.0 ‡‡‡	3.0	rFVIIIFc (Eloctate, Sanofi, Waltham, MA)	83 IU/kg q.d.	581	0	41	NR	NR	59	71	Negative	rFVIIIFc prophylaxis
1‐4 §§§	Intron−22	11.0	11.0	rFVIII (Xyntha, Pfizer, Philadelphia, PA)	50 IU/kg t.i.w.	150	4	64	112	112	64	64	Negative	rFVIIIFc prophylaxis
1‐6	Intron−22	378.7	378.1	rFVIII (Advate, Baxalta US Inc, Lexington, MA)	96 IU/kg q.d.	672	1	99	N/A	N/A	99	157	1.3 ¶¶¶	rFVIIIFc ITI
1‐10 §§	Insertion	28.8	6.2	Missing data	100 IU/kg q.d.	700	6	N/A	N/A	N/A	N/A	59	4.4	rFVIIIFc ITI

Abbreviations: BU, Bethesda unit; FVIII, factor VIII; ITI, immune tolerance induction; N/A, not applicable; NR, not reported; q.d., once daily; rFVIIIFc, recombinant factor VIII Fc fusion protein; t.i.w., three times per week.

Patients are sorted in ascending order according to time from the start of ITI to tolerization. Patient numbers were randomly assigned.

Bolded data indicate high‐risk features.

Time to first negative inhibitor titre: time interval (in weeks) from the start date of ITI treatment with rFVIIIFc to date of the patient's first time reaching inhibitor titre of <0.6 BU/mL.

Time to FVIII normal recovery: time interval (in weeks) from the date of ITI treatment with rFVIIIFc to date of the patient's first time reaching FVIII recovery level of ≥66% of expected.

Time to FVIII half‐life of ≥6 hours: time interval (in weeks) from the start date of ITI treatment with rFVIIIFc to date of the patient's first time reaching FVIII half‐life of ≥6 hours.

Time to tolerization: time interval (in weeks) from the start date of ITI treatment with rFVIIIFc to the date when physician reported this patient reached tolerization.

Newly identified patient.

Received rituximab concomitantly with rFVIIIFc.

This patient was first on rFVIIIFc ITI (83 IU/kg q.d.) for 15 weeks (titre=26 BU/mL), switched away to another factor ITI for 13 weeks and then restarted rFVIIIFc ITI on 29 March 2017 (titre=44 BU/mL) with rFVIIIFc 21 IU/kg per hour drip treatment regimen, and achieved negative inhibitor titre 13 weeks after restart of rFVIIIFc ITI and was tolerized after 32 weeks of treatment; patient is currently on rFVIIIFc prophylaxis.

This patient was enrolled with a historical peak inhibitor titre of 30.0 BU/mL. During the final data cleaning, the value was corrected to be 3.0 BU/mL instead.

This patient transitioned to rFVIIIFc prophylaxis after 64 weeks of rFVIIIFc ITI treatment, laboratory assessments on normal recovery and time to half‐life ≥6 hours were available 58 weeks after the patient transitioned to rFVIIIFc prophylaxis.

This patient was considered tolerized by the treating physician but showed a low‐titre inhibitor during the follow‐up period and remains on rFVIIIFc ITI at the time of data capture.

Table 2

Rescue ITI patients ,

Patient	FVIII genotype	Number of prior ITI regimens	Inhibitor titre (BU/mL)		Factor brand being used when inhibitor developed	rFVIIIFc ITI regimen	Weekly factor usage (IU/kg)					Current titre (BU/mL)	Current status
			Historical peak (pre‐ITI)	Immediately pre‐rFVIIIFc ITI				Inhibitor diagnosis to start of rFVIIIFc ITI	Start of ITI to		Duration of rFVIIIFc ITI
									Negative Bethesda titre §	Tolerization ¶
2‐4 ††	Intron−22	1	1178.0	1.0	rFVIII (Advate, Baxalta US Inc, Lexington, MA)	100 IU/kg q.o.d.	350	94	13	22	22	Negative	rFVIIIFc prophylaxis
2‐1	Intron−22	7	250.0	9.0	rFVIII (Advate, Baxalta US Inc, Lexington, MA)	200 IU/kg q.d.	1400	297	28	35	35	Negative	rFVIIIFc prophylaxis
2‐19 ‡‡ , ¶¶	Intron−22	2	224.0	15.0	rFVIII (Advate, Baxalta US Inc, Lexington, MA)	100 IU/kg q.o.d.	350	238	14	47	80	0.9	rFVIIIFc ITI
2‐9	Intron−22	3	11.0	1.3	rFVIII (Advate, Baxalta US Inc, Lexington, MA)	100 IU/kg q.o.d.	350	626	100	101	135	Negative	rFVIIIFc prophylaxis
2‐2	Intron−22	5	67.0	4.0	rFVIII (Advate, Baxalta US Inc, Lexington, MA)	150 IU/kg q.d.	1050	249	3	N/A	41	7.0	Emicizumab
2‐7 ‡‡	Nonsense mutation	1	306.0	129.0	rFVIII (Advate, Baxalta US Inc, Lexington, MA)	100 IU/kg q.d.	700	243	13	N/A	87	36.0	Emicizumab
2‐5 ‡‡ , †††	Intron−22	2	460.0	200.0	rFVIIIFc (Eloctate, Sanofi, Waltham, MA)	150 IU/kg q.d.	1050	42	13	N/A	90	Negative	rFVIIIFc prophylaxis
2‐3	Partial gene deletion	3	100.0	34.6	rFVIII (Recombinate, Baxalta US Inc, Lexington, MA)	191.5 IU/kg q.o.d.	670	498	31	N/A	82	14.6	rFVIIIFc ITI; BPA prophylaxis
2‐6	Intron−22	3	41.8	22.3	rFVIII (Advate, Baxalta US Inc, Lexington, MA)	130 IU/kg q.d.	910	265	68	N/A	169	2.4	Emicizumab
2‐10	Intron−22	2	8.0	0.6	rFVIII (Advate, Baxalta US Inc, Lexington, MA)	100 IU/kg q.3.d.	233	439	70	N/A	83	Negative	rFVIIIFc ITI
2‐8	Inversion	1	43.7	35.6	rFVIII (Advate, Baxalta US Inc, Lexington, MA)	200 IU/kg q.o.d.	700	271	N/A	N/A	68	44.0	rFVIIIFc ITI
2‐11	Large deletion	4	1024.0	237.0	rFVIII (Helixate, CSL Behring LLC, Kankakee, IL)	100 IU/kg q.d.	700	473	N/A	N/A	38	1024.0	rFVIIIFc ITI
2‐12	Nonsense mutation	4	409.0	26.0	rFVIII (Helixate, CSL Behring LLC, Kankakee, IL)	100 IU/kg q.d.	700	491	N/A	N/A	94	166.0	BPA prophylaxis
2‐13 ¶¶	Insertion	6	18.0	1.9	rFVIII (Refacto, Wyeth, Philadelphia, PA)	130 IU/kg q.d.	910	989	N/A	N/A	47	5.0	Emicizumab
2‐14 ¶¶	Unknown	1	29.0	27.2	Missing data	43 IU/kg t.i.w.	129	2242	N/A	N/A	70	2.5	rFVIIIFc ITI
2‐15 ¶¶	Intron−22	2	24.0	4.1	rFVIII (Kogenate, Bayer HealthCare LLC, Whippany, NJ)	52 IU/kg t.i.w.	156	934	N/A	N/A	33	40.6	BPA prophylaxis
2‐16 ¶¶	Unknown	1	110.0	50.0	rFVIII (Advate, Baxalta US Inc, Lexington, MA)	186 IU/kg q.d.	1302	32	N/A	N/A	32	26.2	rFVIIIFc ITI
2‐17 ¶¶	Small deletion	2	410.0	99.2	rFVIII (Kogenate FS, Bayer HealthCare LLC, Whippany, NJ)	200 IU/kg q.d.	1400	216	N/A	N/A	11	72.0	Humate‐P prophylaxis; BPA prophylaxis
2‐18 ¶¶	Intron−22	3	275.0	1.0	rFVIII (Kogenate, Bayer HealthCare LLC, Whippany, NJ)	100 IU/kg q.o.d.	350	467	N/A	N/A	24	34.8	Emicizumab

Abbreviations: BPA, bypass agent; BU, Bethesda unit; FVIII, factor VIII; ITI, immune tolerance induction; N/A, not applicable; q.3.d., every three days; q.d., once daily; q.o.d., every other day; rFVIIIFc, recombinant factor VIII Fc fusion protein; t.i.w., three times per week.

Patients are sorted in ascending order according to time from the start of ITI to tolerization first and then to negative Bethesda titre. Patient numbers were randomly assigned.

Bolded data indicate high‐risk features.

Time to first negative inhibitor titre: time interval (in weeks) from the start date of ITI treatment with rFVIIIFc to date of the patient's first time reaching inhibitor titre of <0.6 BU/mL.

Time to tolerization: time interval (in weeks) from the start date of ITI treatment with rFVIIIFc to the date when the physician reported that this patient reached tolerization.

This patient stopped traditional ITI after 21.7 weeks of rFVIIIFc ITI treatment and transitioned to enhanced rFVIIIFc prophylaxis.

Received rituximab concomitantly with rFVIIIFc.

This patient was tolerized after 47 weeks of rFVIIIFc ITI treatment and re‐developed inhibitors approximately 10 weeks after tolerization.

Newly identified patient.

Patient reached negative Bethesda titre 13 weeks after the start of rFVIIIFc ITI; stopped rFVIIIFc ITI with BU=2, switched to another factor ITI and tolerized; now this patient is on rFVIIIFc prophylaxis (116 IU/kg q.o.d.).

First‐time ITI patients had median (range) historical peak inhibitor titre of 45.1 (3.0–1126.0) BU/mL and median (range) time from inhibitor diagnosis to start of rFVIIIFc ITI of 6.4 (0.0–41.0) weeks. Median (range) inhibitor titre at start of rFVIIIFc ITI was 28.8 (3.0–1126.0) BU/mL. Dosing regimens for rFVIIIFc ITI varied; median (range) dose was 100 (50–200) IU/kg and median (range) weekly dose was 700 (150–1400) IU/kg. One first‐time ITI patient received rituximab during rFVIIIFc ITI.

Rescue ITI patients had median (range) historical peak inhibitor titre of 110.0 (8.0–1178.0) BU/mL, median (range) time from inhibitor diagnosis to start of rFVIIIFc ITI of 296.9 (31.6–2242.4) weeks (5.7 [0.6–43.0] years), had undergone a median (range) of 2 (1–7) prior ITI courses and had median (range) inhibitor titre at start of rFVIIIFc ITI of 22.3 (0.6–237.0) BU/mL. Dosing regimens for rFVIIIFc ITI varied; median (range) dose was 100 (43–200) IU/kg and median (range) weekly dose was 700 (129–1400) IU/kg. Three rescue patients received rituximab during rFVIIIFc ITI.

Nine out of 10 patients receiving first‐time ITI using rFVIIIFc (including the patient who received rituximab) achieved a negative Bethesda titre at a median (range) of 30 (3–99) weeks (mean [standard deviation (SD)]: 34.0 [31.2] weeks), achieved tolerance at a median (range) of 30 (3–99) weeks (mean [SD]: 41 [29] weeks) and 8 transitioned to rFVIIIFc prophylaxis. One patient who achieved Bethesda negativity and was considered by their physician to be tolerized showed a low‐titre inhibitor (1.3 BU/mL) during the follow‐up period; this patient remained on rFVIIIFc ITI at the time of data capture. The tenth patient had a decreased Bethesda titre from 6.2 BU/mL at the start of rFVIIIFc to 4.4 BU/mL at 59 weeks and continued on rFVIIIFc ITI.

Over half (10/19) of the patients receiving rescue ITI reached a negative Bethesda titre after a median (range) of 21 (3–100) weeks (mean [SD]: 35.3 [32.6] weeks); 4 were subsequently tolerized (at 22, 35, 47 and 101 weeks; 3 of these transitioned to rFVIIIFc prophylaxis and 1 relapsed and returned to rFVIIIFc ITI), 3 were on emicizumab at the time of data capture, 1 was tolerized on another FVIII product and afterwards transitioned to rFVIIIFc prophylaxis and 2 continued rFVIIIFc ITI. Of the 9 rescue patients who had not reached a negative Bethesda titre at the time of data capture, 4 remained on rFVIIIFc ITI; 5 stopped rFVIIIFc ITI and transitioned to either emicizumab (n = 2), prophylaxis with a bypass agent (n = 2) or prophylaxis with another FVIII replacement therapy and bypass agent (n = 1).

Altogether, 24/29 patients (9 first‐time, 15 rescue) had a central venous access device in place before commencing rFVIIIFc ITI. Most patients (19/29 [66%]: 9 first‐time, 10 rescue) began rFVIIIFc ITI on a daily dosing regimen, ranging from 83 to 200 IU/kg daily. Twelve (41%) patients changed their ITI dosing regimen at some point. Most patients (23/29 [79%]) did not report any adherence issues. At the time of data capture, 21/29 patients (72%; 10/10 first‐time, 11/19 rescue) were receiving rFVIIIFc (prophylaxis or ITI). One rescue patient received bypass agent prophylaxis in addition to rFVIIIFc ITI.

No adverse events were assessed as related to rFVIIIFc. In total, 19 surgeries were performed concomitant with ITI (eight [two major and six minor] in first‐time and 11 [10 minor and one unclassified] in rescue patients). The two major surgeries were craniotomy and reconstruction of a left parietal defect in 2 patients. rFVIIIFc ITI was uninterrupted during all surgery and post‐operative periods; bypass agent–controlled bleeding during all procedures among first‐time patients and 7/11 procedures among rescue patients.

This retrospective chart review in a real‐world setting shows that first‐time ITI patients achieved rapid tolerization with a high success rate (80%) using rFVIIIFc. Among rescue patients, more than half reached a negative titre within 21 weeks of starting rFVIIIFc ITI and 4 subsequently reached tolerization. This was achieved using various dosing regimens with lower factor usage than recommended to date for success in this high‐risk group.

The results demonstrate a shorter median time to tolerization with rFVIIIFc ITI than reported with other FVIII regimens or with von Willebrand factor‐containing plasma‐derived FVIII. Despite being at a higher risk of ITI failure and receiving half of the median factor dose (700 vs 1400 IU/kg/week) administered to patients in the high‐dose arm of the International Immune Tolerance study, this population took markedly less time to achieve tolerance than in that study.

Our results match previous observations that achieving successful tolerization in rescue ITI patients is generally difficult and much less likely to be successful, making the first attempt at ITI most important. Increasingly, as well, clinicians advocate for commencing ITI as soon as possible after high‐titre inhibitor development. Our analysis showed that, for the most part, clinicians involved in this North American real‐world study started ITI (in first‐time ITI patients) without waiting for inhibitor titres to drop to a predefined level. Supporting this approach, all first‐time ITI patients initiating rFVIIIFc ITI within 1 month of inhibitor diagnosis were tolerized.

The high success rate among patients undergoing first‐time ITI included in this chart review may be due partly to potential immunomodulatory properties of rFVIIIFc. Further study of the immunogenicity of rFVIIIFc, in previously untreated patients with haemophilia A, is being analysed (ClinicalTrials.gov: NCT02234323).

Limitations of this study include its retrospective nature, small patient population and potential for reporting biases. The impact of ITI initiation soon after inhibitor detection is not fully understood and may have contributed to the success of first‐time ITI. Additionally, the definition of tolerization applied in this study included attaining a 6‐hour FVIII half‐life. While this has been an accepted parameter for characterizing tolerization in an era of extended half‐life factors, new studies are required to determine the appropriate half‐life target for defining success of ITI.

Although the haemophilia treatment landscape is changing with the advent of emicizumab as well as potentially other rebalancing therapies, all of which can be used in patients with inhibitors, eradication of inhibitors remains an important goal for patients with high‐titre inhibitors and ITI continues to be the standard of care for these patients. However, current ITI regimens require frequent factor infusions and a long duration of treatment, and are only efficacious in 50%–70% of patients. More effective regimens that establish Bethesda negativity and achieve successful ITI more quickly would likely reduce the substantial risk of bleeding during early ITI (this may be mitigated by concomitant administration of emicizumab during ITI), improve long‐term patient outcomes and reduce treatment burden and improve patient quality of life. Since ITI is typically costly, more effective and efficient tolerization could also reduce healthcare utilization and costs associated with ITI.

In conclusion, extended half‐life rFVIIIFc is an effective option for ITI therapy in patients with severe haemophilia A and inhibitors at high risk of ITI failure in a real‐world setting. Prospective studies are underway assessing the efficacy of first‐time and rescue rFVIIIFc ITI in patients with haemophilia A who have developed inhibitors (verITI‐8 [NCT03093480]; reITIrate [NCT03103542]).

CONFLICT OF INTEREST

The views expressed in this manuscript are those of the authors and do not reflect the official policy of the Department of the Army/Navy/Air Force, Department of Defense or US Government. Manuel Carcao has received research support from Bayer, Bioverativ, a Sanofi company, CSL Behring, Novo Nordisk, Octapharma, Pfizer and Shire and participated in speaker/advisory boards for Bayer, Sanofi, Biotest, CSL Behring, Grifols, LFB, Novo Nordisk, Octapharma, Pfizer, Roche and Shire. Amy Shapiro has nothing to disclose. Nina Hwang is a consultant for Bayer, Shire, HEMA Biologics, BP and is a principal investigator for Sanofi. Steven Pipe is a consultant for Takeda, Novo Nordisk, Sanofi, CSL Behring, Pfizer, Genentech/Roche, Alnylam, ApcinteX, BioMarin, uniQure, Bayer, Freeline, Spark Therapeutics, Catalyst Biosciences and HEMA Biologics and has received research funding from Shire and Siemens. Sanjay Ahuja has received honoraria from Bayer, Shire and Sanofi and has participated in a speakers’ bureau for Shire and Sanofi. Ken Lieuw has nothing to disclose. Janice Staber received honoraria from Genentech, Sanofi, BioMarin, Novo Nordisk, Spark Therapeutics and Bayer. Mark Belletrutti has received honoraria from Sanofi, Octapharma, Roche and Takeda, received research support from Sanofi, Octapharma and Takeda and received investigator‐initiated study funding from Octapharma. Haowei Linda Sun has participated in advisory boards for Novo Nordisk and Octapharma and has received research funding from Octapharma. Hilda Ding is a principal investigator for Sanofi. Michael Wang is a consultant or advisor for Bayer, Sanofi, CSL Behring, Novo Nordisk, Takeda, Genentech, Catalyst Biosciences and BioMarin. Victoria Price has received research funding from Shire, Novo Nordisk and Sanofi and sat on advisory boards for Shire, Roche, Sanofi and Bayer. MacGregor Steele received honoraria from Baxter/Shire, Roche and Bayer. Elisa Tsao is an employee of Sanofi. Jing Feng was an employee of Sanofi at the time of the study. Zahra Al‐Khateeb is a former employee of Trinity Partners LLC, a consulting firm retained by Sanofi, to conduct the study on which this manuscript is based. Jennifer Dumont is an employee of and holds equity interest in Sanofi. Nisha Jain was an employee of Sanofi at the time of the study.

AUTHOR CONTRIBUTIONS

M. Carcao, E. Tsao, J. Feng, J. Dumont and N. Jain were responsible for the study concept and design. M. Carcao, A. Shapiro, N. Hwang, S. Pipe, S. Ahuja, K. Lieuw, J. Staber, M. Belletrutti, H. L. Sun, H. Ding, M. Wang, V. Price, M. Steele and Z. Al‐Khateeb were responsible for data acquisition. All authors contributed to the interpretation of data, writing and revising the letter, as well as providing final approval of the version to be published.