Optimising prophylaxis outcomes and costs in haemophilia patients switching to recombinant FVIII-Fc: a single-centre real-world experience

Abstract

Background

The recombinant factor VIII (rFVIII)-IgG1 Fc fusion protein (rFVIII-Fc) was the first available extended half-life rFVIII, shown to prolong dosing intervals of individualised prophylaxis in patients with severe haemophilia A, maintaining low bleeding rates and unchanged or lower FVIII dose versus standard half-life (SHL) rFVIII. Few data are available about real-world experience with rFVIII-Fc, including criteria for patient switching from SHL products, follow up and prophylaxis optimisation.

Materials and methods

A single-centre retrospective study was designed to review patients switched to rFVIII-Fc, based on individual needs, after pharmacokinetic (PK) assessment, according to routine clinical practice. In patients with adequate post-switch follow up, data about rFVIII-Fc prophylaxis were compared with those from the last 18-months SHL rFVIII prophylaxis.

Results

Of 25 candidates, 18 patients (15 severe, 3 moderate; aged 9–62 years; 3 with inhibitor history) started rFVIII-Fc regimens, with comparable FVIII weekly dose and reduced infusion frequency (mean −30%) in all 17 patients previously on SHL rFVIII prophylaxis thrice weekly or every other day. Over a mean 18-month follow up in 13 patients, compared with SHL products, further reduced infusion frequency (mean −40%; p<0.001; interval ≥4 days in 9 patients), improved treatment satisfaction (Hemo-sat questionnaires), significantly lower FVIII weekly dose and annual consumption (mean −12%; p=0.019), comparable bleeding rates and FVIII trough levels, and improved management of breakthrough bleeding were observed. von Willebrand Factor Antigen (VWF:Ag) correlated to PK variables and both had relationships with rFVIII-Fc weekly dose, increasing statistical significance over the follow-up period. No inhibitors or drug-related adverse events were recorded.

Discussion

In this real-world series of patients, a switch to rFVIII-Fc, based on careful assessment of clinical needs, PK testing and treatment monitoring, was able to optimise individual convenience, efficacy and costs of prophylaxis.

INTRODUCTION

There is consistent evidence of the benefits of prophylaxis, i.e. the regular, long-term intravenous (IV) administration of factor concentrates to prevent bleeding and its deleterious effects into joints in persons with severe haemophilia (factor VIII/IX, <1%) of all ages^1–4. The reduction in bleeding frequency and in the impact of the disease on joint status and quality of life make prophylaxis the most adopted treatment regimen in all age groups, at least in many high-income countries⁵. However, the need for frequent venous access (up to every 2 days in haemophilia A) remains the main barrier to implementation of prophylaxis in children and to long-term adherence to treatment at any age^6,7. Further challenges concern the optimisation of prophylaxis in the presence of breakthrough bleeding or specific clinical needs, including target joints and evolving arthropathy, patients’ lifestyle and physical activity^7,8. Personalised prophylaxis is, therefore, increasingly addressed in clinical practice, even to take into account the large variability in the individual pharmacokinetic (PK) responses to the different products^9,10. In this scenario, the availability of factor concentrates with extended half-life (EHL) has been eagerly awaited, with the aim of reducing the burden of venous injections and improving clinical outcomes. However, in clinical trials, irrespective of the technology employed, modified FVIII concentrates showed an approximately 1.5-fold prolongation of half-life. This was below expectations, particularly when compared with the 5-fold extension provided by new FIX products¹¹.

Efmoroctocog alfa (Elocta^®, Swedish Orphan Biovitrum, Stockholm, Sweden), a recombinant B-domain deleted FVIII linked to the Fc fragment of human immunoglobulin (Ig) G1 (rFVIII-Fc), was the first available EHL FVIII product¹² and is, to date, the only one marketed in Italy¹³. Thanks to the binding to the neonatal Fc receptor and the Ig recycling pathway¹⁴, rFVIII-Fc showed a terminal half-life of 19 h (geometric mean) in the phase III study in severe haemophilic patients aged ≥12 years, who achieved low annualised bleeding rates (ABR) on individualised prophylaxis; the dosing interval was extended to ≥3 days in almost all patients¹⁵. These benefits also applied to children¹⁶ and were maintained in the extension study, in which most patients did not require adjustment of prophylaxis regimens and some (22% adolescents/adults, 3% children) further prolonged dosing intervals, with unchanged, or even lower, prophylactic FVIII dose¹⁷.

Despite these encouraging results, uncertainties remain among physicians and patients about any significant advantages and safety in switching to EHL FVIII products¹⁸. In fact, there is little real-world evidence available^19–21 and information about minimally treated/previously unexposed patients or those with inhibitor history is lacking.

Here we report data on switching from standard half-life (SHL) FVIII products, follow up and optimisation of prophylaxis regimens with rFVIII-Fc in a series of consecutive patients in the clinical practice at an Italian haemophilia centre.

MATERIALS AND METHODS

This study was a retrospective review of patients treated with rFVIII-Fc since its market availability up to March 2019 at the Regional Reference Centre for Inherited Bleeding Disorders, University Hospital of Parma, Italy. In patients with adequate post-switch follow up, data about rFVIII-Fc prophylaxis were compared with those from the last 18-months prophylaxis with SHL FVIII concentrates (i.e. a time period comparable to the mean patients’ follow up on rFVIII-Fc). The study protocol was approved by the Institutional Ethics Review Board (registration no. 3658/2019).

Subjects

All patients who received rFVIII-Fc were eligible and were enrolled after signing a specific informed consent. According to the Centre’s clinical practice, switching from SHL products to rFVIII-Fc was evaluated in patients with severe or moderate haemophilia without inhibitors on or with indication to regular prophylaxis at follow-up visits, considering the possible advantages and risks. After the shared decision with patients (or parents/guardians of minors) and informed consent to changing FVIII concentrate according to national recommendations²², patients underwent a single-dose rFVIII-Fc PK study with careful clinical and laboratory follow up, once they had started regular prophylaxis with rFVIII-Fc (see below). According to the Centre’s standard approach, switching was not considered in patients unable to undergo the PK assessment and the post-switch follow up. In the absence of adherence or venous access problems, and in some patients with good outcomes on twice weekly prophylaxis regimens, the issue of switching was not addressed unless requested by the patient. Moreover, patients with a history of high-responding inhibitors and difficulty in their eradication (prolonged treatment, need for rescue) after immune tolerance induction (ITI), irrespective of duration of tolerance, were excluded.

Pharmacokinetics assessment

In accordance with expert recommendations^10,13,23, after at least 3 days wash-out, in a non-bleeding state, 45–50 IU/kg rFVIII-Fc was administered over <10 min. Blood samples for FVIII coagulant activity (FVIII:C) were collected at baseline, 15 and 60 min after the end of infusion, and 3, 6, 24, 48, 72 and 96 h thereafter. Baseline sampling included anti-FVIII inhibitor and von Willebrand Factor Antigen (VWF:Ag) assessments.

Non-compartmental PK analyses were performed using the Phoenix WinNonlin software package (Pharsight Corp., Mountain View, CA, USA). Calculated parameters included terminal half-life (HL), area under the curve from the time of infusion to the last measurement (AUC), clearance (normalised by body weight), and mean residence time (MRT). Incremental in vivo recovery (IVR) was calculated by dividing the maximum observed FVIII:C level and the infused dose.

Prophylaxis regimens and clinical and laboratory follow up

Individualised prophylaxis regimens with rFVIII-Fc were based on PK assessment, to achieve FVIII trough levels comparable to those previously reported on SHL rFVIII, and at least >1%. However, first consideration was given to specific clinical needs related to joint status, physical activity or infusion burden reduction.

Post-switch follow up, according to the Centre’s standard practice, included outpatient visits and laboratory investigations monthly over the first 3 months and every 3 months thereafter, for up to one year. Subsequently, the routine 6-month follow up was resumed. All visits were scheduled at the longest prophylaxis dosing interval to perform blood sampling for FVIII trough level measurement.

Data collection and assessment

Patients’ clinical characteristics, including severity of disease, age, inhibitor history, duration and type of prophylaxis, were obtained through the institutional web-based clinical record “xl’Emofilia”²⁴, which also provided detailed information about breakthrough bleeds (type and severity, spontaneous/post-traumatic), FVIII trough levels and FVIII concentrate administration, registered by physicians (hospital infusions) or patients (home treatment, with physicians’ validation).

Efficacy outcomes included ABR (with emphasis on spontaneous bleeds, AsBR), FVIII trough levels, and adherence to treatment (defined as the ratio between registered infusions and those planned according to the prescribed prophylaxis regimen). FVIII concentrate consumption and number of injections on prophylaxis were calculated for the last 18-months pre-switch and for rFVIII-Fc regimens (last prescribed dosing and interval).

Safety records included anti-FVIII inhibitor monitoring and adverse events related to treatment, if any.

Data addressing treatment satisfaction before switch and after at least 6 months of rFVIII-Fc prophylaxis were available from the validated specific Hemo-sat questionnaire²⁵, consisting of 34 items pertaining to six dimensions (“ease and convenience”, “efficacy”, “burden”, “specialist”, “centre” and “general satisfaction”) routinely adopted at the Centre for monitoring patients changing product/regimen of treatment. In all cases, the lower the score, the higher the level of treatment satisfaction.

Measurements

Available FVIII:C were obtained using a one-stage coagulation assay (HemosIL^® SynthASil, Instrumentation Laboratory/Werfen, Bedford, MA, USA). Samples from each PK assessment were measured in single laboratory sessions. Mean FVIII trough levels from the last two assessments on pre-switch and the last prescribed rFVIII-Fc regimens were reported. Inhibitor was tested by the Bethesda method with Nijmegen modification, using the same FVIII assay. VWF:Ag was measured by an automated latex immunoturbidimetric assay (HemosIL^® VWF:Ag, Instrumentation Laboratory/Werfen).

Statistical analysis

Descriptive statistics report numbers and percentages for categorical variables and mean and 1 standard deviation (SD) for continuous variables. Differences in categorical variables were evaluated by χ² statistics, whereas the Student t-test was used for continuous variables, for paired or independent samples as appropriate. Bivariate correlations were expressed by the Pearson’s coefficient (r). All analyses were performed by the SPSS ver. 22 (SPSS Statistics, IBM, Armonk, NY, USA). p<0.05 was considered statistically significant.

RESULTS

Patient series

Among subjects with severe (n=34) or moderate (n=8; FVIII:C 1.0–2.8%) haemophilia A on (all severe patients but one and 5 of 8 moderate patients) or with indications for regular prophylaxis, 25 potential candidates were identified (Online Supplementary Content, Figure S1). Switch was declined by one moderate patient treated on-demand, because of fear of inhibitor development, despite negative personal and family history. Perplexities and concerns of 4 severe patients delayed further assessments.

Overall, 20 patients underwent rFVIII-Fc PK assessment. Their demographic and clinical characteristics, with reason(s) for switching, are reported in Table I. Patients were all Caucasian males, aged 30.5±15.5 years (mean ± 1 SD), all but one on long-term prophylaxis (duration 14.2±4.8 years) started at 18.4±15.6 years of age with SHL recombinant FVIII products (octocog alfa from BHK and CHO cell lines and moroctocog alfa). Mean dosing on prophylaxis was 31.2±5.8 IU/kg, thrice weekly in 12 patients and every other day (e.o.d) in 6. Four patients had inhibitor history, one a low-titre (peak 2.4 BU/mL) transient inhibitor and three high-titre inhibitors eradicated by ITI 6–12 years earlier.

Table I.

Patient’s clinical characteristics, regimens on standard half-life (SHL) rFVIII prophylaxis and reasons to switch to rFVIII-Fc with first prescribed regimen

Patient	FVIII:C (%)	Age (years)	Inhibitor history	Type of prophylaxisa	Age at start of prophylaxis (years)	Prophylaxis duration (years)	SHL rFVIII prophylaxis regimen	Reason(s)b to switch	rFVIII-Fc prophylaxis regimen
1	<1	29	No	Secondary	9.9	19.2	33 IU/kg×3/wk	Reduce infusion n (VAP)	51 IU/kg×2/wk
2	<1	34	No	Secondary	12.7	21.5	32 IU/kg×3/wk	Reduce infusion n	50 IU/kg×2/wk
3	1.1	30	No	Secondary	10.7	19.3	30 IU/kg×3/wk	Reduce infusion n	42 IU/kg×2/wk
4	<1	45	No	Tertiary	30.3	14.8	37 IU/kg e.o.d.	Reduce infusion n, higher protection (ART, PA)	47 IU/kg×2/wk
5	<1	16	Yesc	Secondary	1.7	14.7	35 IU/kg e.o.d.	Higher protection (PA), reduce infusion n	46 IU/kg every 3 days
6	<1	56	No	Tertiary	41.9	14.7	37 IU/kg×3/wk	Reduce infusion n (ADH)	50 IU/kg×2/wk
7	2	21	No	Secondary	8.8	13.1	24 IU/kg×3/wk	Reduce infusion n, higher protection	46 IU/kg×2/wk
8	<1	19	No	Secondary	3.4	16	25 IU/kg×3/wk	Reduce infusion n 47	IU/kg×2/wk
9	<1	35	No	Tertiary	27	8.3	34 IU/kg×3/wk	Reduce infusion n, higher protection (PA)	45 IU/kg every 3 days
10	<1	21	No	Primary	1.1	20.8	23 IU/kg×3/wk	Reduce infusion n (VAP)	44 IU/kg×2/wk
11	<1	62	No	Tertiary	52	10.8	23 IU/kg×3/wk	Reduce infusion n (ADH)	36 IU/kg every 4 days
12	<1	39	No	Tertiary	18.3	21.3	40 IU/kg×3/wk	Reduce infusion n (ADH), higher protection (ART)	50 IU/kg every 3 days
13	<1	15	Yesd	Secondary	1.3	14	27 IU/kg e.o.d.	Reduce infusion n, higher protection (PA)	45 IU/kg every 3 days
14	<1	37	Yesd	Tertiary	26	12	25 IU/kg×3/wk	Reduce infusion n	–g
15	<1	33	No	Tertiary	20	13.3	37 IU/kg×3/wk	Reduce infusion n	51 IU/kg every 3 days
16	<1	49	No	Tertiary	40	9.1	40 IU/kg e.o.d.	Reduce infusion n, higher protection (ART)	52 IU/kg every 3 days
17	2	9	No	–	–	–	–	Prophylaxis feasibilitye	48 IU/kg×2/wk
18	<1	12	Yesd	Secondary	5.5	6.5	42 IU/kg e.o.d.	Reduce infusion n (VAP), higher protection (PA)	52 IU/kg×3/wk
19	<1	10	No	Primary	3	7	60 IU/kg e.o.d.	Reduce infusion n (VAP), higher protection (PA)	55 IU/kg×3/wk
20	<1	19	No	–	–	–	–	Prophylaxis feasibilityf	–h

^aAccording to the definitions by the World Federation of Hemophilia².

^bIn order of clinical relevance in each patient.

^cLow-titre transient inhibitor.

^dHigh-titre inhibitor, eradication by immune tolerance induction.

^eIn this boy with a high bleeding rate, prophylaxis was unfeasible due to poor peripheral vein access and family problems in addressing alternative management and need for regular treatment at the Centre.

^fThis young severe patient started secondary prophylaxis in several occasions but treatment was withdrawn because of intolerance to frequent venipunctures and poor family compliance to clinical follow up.

^gThis severe patient remained on the SHL concentrate used, no relevant advantages being predicted by rFVIII-Fc PK assessment (terminal HL 13 hours, similar 72-hour FVIII:C level than that on current product, after higher concentrate dose).

^hThe patient agreed to undergo rFVIII-Fc PK assessment but did not complete the study.

ADH: adherence; ART: arthropathy; e.o.d.: every other day; n: number; PA: physical activity; rFVIII: recombinant factor VIII; rFVIII-Fc: recombinant FVIII-Fc fusion protein; SHL: standard half life; VAP: venous access problems; wk: week.

As far as reasons for switching to rFVIII-Fc were concerned, in all patients reducing the IV infusion burden was considered, particularly to overcome venous access problems or improve adherence. In 9 patients, a potentially higher protection from bleeding risk, due to physical activity or to joint status, was also addressed.

Eighteen patients started rFVIII-Fc prophylaxis regimens, based on individual PK profiles and clinical needs (Table I), allowing reduced infusion frequency (mean −30%) in all 17 patients previously on SHL rFVIII prophylaxis (Table II, panel A). A thrice weekly regimen was needed only in 2 children, while most patients (n=9) started rFVIII-Fc prophylaxis twice weekly; further 6 patients received rFVIII-Fc every 3 days and one every 4 days.

Table II.

Changes of prophylaxis infusion regimens (A) from standard half-life (SHL) rFVIII concentrates to the first prescription of rFVIII-Fc in 18 patients and (B) to the last prescribed regimen in 13 patients over the follow up

A.	rFVIII-Fc first regimen (n=18)					B.	rFVIII-Fc last prescribed regimen (n=13)
SHL rFVIII regimen (patients n)	e.o.d.	3×wk	Every 3 days	2×wk	Every 4 days	rFVIII-Fc first regimen (patients n)	Every 3 days	2×wk	Every 4 days	Every 5 days	Every 6 days
e.o.d. (n=6)	-	2a	3	1	-	Every 3 days (n=4)	1	2	1	-	-
3×wk (n=11)	-	-	3	7	1	2×wk (n=8)	-	1	6	1	-
On-demand (n=1)	-	-	-	1	-	Every 4 days (n=1)	-	-	-	-	1

^aTwo children (aged 10 and 12 years). All 17 patients on prophylaxis regimens prolonged the dosing interval switching from SHL rFVIII concentrates to rFVIII-Fc (A). The infusion frequency was further reduced in 11 of 13 patients over the follow-up period (B). Grey cells represent patients maintaining their infusion regimen. e.o.d.: every other day; n: number; rFVIII: recombinant factor VIII; rFVIII-Fc: recombinant FVIII-Fc fusion protein; SHL: standard half life; wk: week.

Two severe patients continued treatment with their SHL rFVIII products. One young patient did not complete the PK study (#20, Table I). One adult (#14), who aimed to reduce the infusion burden of his thrice weekly prophylaxis, had no relevant advantages predicted by the individual PK (terminal HL 13 h and 72-h FVIII:C similar to that that on the rFVIII in use, after higher rFVIII-Fc dosing). Five patients (#15 to #19) were on treatment with rFVIII-Fc for ≤3 months; their follow up was considered inadequate to assess outcomes and allow any comparison with previous SHL rFVIII prophylaxis. These study objectives were carefully addressed in 13 patients receiving rFVIII-Fc for 18.2±6.0 months (Table III).

Table III.

Regimen adjustment and outcomes of prophylaxis in 13 patients switched to rFVIII-Fc, compared with their previous standard half-life (SHL) rFVIII prophylaxis.

Patient	SHL rFVIII prophylaxis					rFVIII-Fc prophylaxis
	Weekly SHL rFVIII dose (IU/Kg)	Meana FVIII trough (%)	ABR/AsBRb	Injection n/yrb	FVIII (IU) consumption/yrb	Treatment duration, mo.	ED	rFVIII-Fc regimen change(s)c	Last dosing Regimen	Weekly rFVIII-Fc dose, first/lastd (IU/Kg)	Meana FVIII trough (%)	ABR/AsBRe	Injection n/yrf	FVIII (IU) consumption/yrf
1	99	1.1	1.3/0.6	156	312,000	28	239	No	51 IU/kg× 2/wk	102/102	2.3	0.4/0	104	338,000
2	96	1.8	0/0	156	312,000	26	200	Yes, interval	50 IU/kg every 4 days	100/88	3.3	0/0	91	273,000
3	90	5.3	0.7/0	156	312,000	27	208	Yes, interval	42 IU/kg every 4 days	84/74	5.7	0.4/0	91	273,000
4	130	6.8	0.7/0	183	549,000	19	130	Yes, interval	47 IU/kg every 4 days	94/82	6.5	0.6/0	91	364,000
5	123	1.8	1.3/0	162	486,000	20	157	Yes, interval	46 IU/kg every 4 days	107/81	2.6	0/0	91	364,000
6	111	5.5	1.3/0	156	468,000	19	147	Yes, interval	50 IU/kg every 4 days	100/88	2.4	0.6/0	91	364,000
7	72	4.2	0.7/0.7	156	321,000	18	121	Yes, interval	46 IU/kg every 5 days	92/64	4.9	0.6/0	73	219,000
8	75	2.3	0/0	156	260,000	18	141	Yes, interval	47 IU/kg every 4 days	94/82	2.6	0/0	91	273,000
9	102	1.0	0/0	156	416,000	14	128	Yes, interval and dosing	54 IU/kg × 2/wk	105/108	1.0	0/0	104	442,000
10	69	2.2	0/0	156	312,000	15	117	Yes, interval	44 IU/kg every 4 days	88/77	3.2	0/0	91	319,000
11	69	4.4	0/0	156	312,000	14	77	Yes, interval	36 IU/kg every 6 days	63/42	4.5	0/0	61	183,000
12	120	1.1	2.0/1.3	156	468,000	10	97	No	50 IU/kg every 3 days	116/116	1.5	4.0/1.3	122	427,000
13	95	2.5	3.3/0	183	274,500	9	82	Yes, interval and dosing	50 IU/kg× 2/wk	105/100	1.0	1.5/0	104	286,000
Mean	96.2	3.1	0.87/0.2	161	369,423	18.2	142			96.2/84.9	3.2	0.46/0.1	93	317,308
1 SD	20.9	3.2	0.9/0.4	10	94,542	6.0	49			13.1/19.4	1.7	0.6/0.4	15	75,897
								p vs. SHL rFVIII		0.045g	0.7	0.1	<0.001	0.019

^aMean of the last two available measurements at the longest interval after infusion according to the prophylaxis regimen.

^bReferred to the last 18 months on prophylaxis with SHL rFVIII concentrates.

^cAfter the first prescribed regimen ( see Table I).

^dFirst introduced rFVIII-Fc regimen and the last prescribed over the follow up.

^eFor patients treated < 12 months, annualised bleeding rate was extrapolated.

^fCalculated based on the last rFVIII-Fc prescribed prophylaxis regimen;

^gThe difference was statistically significant when the last prescribed rFVIII-Fc regimen was compared to pre-switch SHL prophylaxis, but not if the first rFVIII-Fc regimen was considered ( p=0.9).

ABR: annual(ised) bleeding rate; AsBR: annual(ised) spontaneous bleeding rate; ED: exposure days; rFVIII: recombinant Factor VIII; rFVIII-Fc: recombinant FVIII-Fc fusion protein; SD: standard deviation; SHL: standard half-life; yr: year; wk: week.

Pharmacokinetics of rFVIII-Fc

Figure 1 shows PK profiles of the 19 patients completing the study, with mean FVIII:C at the different time-points after infusion and calculated PK parameters. Terminal HL ranged 10.6–31.9 h and was shorter in children (13.6±4.7 h) than in adolescent/adult patients (19.9±6.2 h). Overall, after rFVIII-Fc 45–50 IU/kg infusion FVIII:C was >5% at 48 h in all patients except 3 (2 children) and remained >3% and >2% at 72 and 96 h, respectively, in 12 patients. Only the 2 above mentioned children showed FVIII:C <1% 72 h after infusion; however FVIII:C was >2% at 48 h (<1% on previous e.o.d. SHL rFVIII).

Figure 1*.

Single-dose rFVIII-Fc pharmacokinetic profiles of the 19 patients who completed the study

All individuals received 45–50 IU/kg rFVIII-FC in non-bleeding state after at least 3-day wash out. FVIII:C were measured by one-stage assay at baseline, 15 minutes (min), 60 min, and 3, 6, 24, 48, 72 and 96 hours after the end of infusion. Mean FVIII:C levels with minimum and maximum are reported. Pharmacokinetics parameters were calculated by the WinNonlin softaware package. AUC: area under the curve; HL: half-life; IVR: in vivo recovery; MRT: mean residence time; rFVIII-Fc: recombinant FVIII-Fc fusion protein. *The coloured figure is published online.

Individual PK variables of the 13 patients with follow up on rFVIII-Fc were analysed in detail (Table IV), including their VWF:Ag levels. Terminal HL was directly correlated with MRT and AUC, and all these variables showed inverse relationships with clearance. Statistically significant correlations were found between VWF:Ag levels and MRT, AUC (both direct) and clearance (inverse; Table IV, last row).

Table IV.

Pharmacokinetic parameters, von Willebrand Factor levels and significant correlations in 13 patients on prophylaxis with rFVIII-Fc

Patient	Terminal HLa,b (hours)	Incremental IVR (IU dL−1/IU kg−1)	AUC (U*h/dL)	MRTb (hours)	Clearance (mL/h/kg)	VWF:Agc (%)
1	15.16	4.13	2,250	14.07	2.05	74
2	17.32	2.37	2,252	19.12	2.16	113
3	25.11	2.26	2,418	26.66	1.59	78
4	25.56	2.12	2,701	26.72	1.57	114
5	18.24	2.60	2,933	21.23	1.87	109
6	17.43	2.32	2,567	21.09	1.91	128
7	27.04	2.57	3,361	27.63	1.22	197
8	17.71	2.15	2,439	21.23	1.53	138
9	15.37	1.51	1,061	16.16	4.15	59
10	31.02	2.25	1,652	20.56	2.49	56
11	31.91	2.86	4,416	30.78	0.83	247
12	13.25	1.96	1,523	15.29	3.26	71
13	18.73	1.97	1,642	17.16	2.69	72
Mean	20.47	2.36	2,343	20.87	2.17	109.1
1 SD	6.41	0.61	865	5.33	0.89	55.3
Significant correlationsr (p)d	AUC 0.57 (0.04) Clearance −0.58 (0.04) MRT 0.82 (0.001)		HL 0.57 (0.04) Clearance −0.91 (<0.001) MRT 0.81 (0.001) VWF:Ag 0.92 (<0.001)	HL 0.82 (0.001) AUC 0.81 (0.001) Clearance −0.78 (0.002) VWF:Ag 0.75 (0.003)	HL −0.58 (0.04) AUC −0.91 (<0.001) MRT −0.78 (0.002) VWF:Ag −0.73 (0.005)	AUC 0.92 (<0.001) MRT 0.75 (0.003) Clearance −0.73 (0.005)

^aFour patients (#1, #4, #11 and #13) had pre-switch SHL rFVIII data available, being HL ratios 1.40, 1.79, 1.53 and 1.59, respectively.

^bIn patients in whom dosing interval was prolonged ≥4 days (n=9) compared to those remaining on twice weekly or every 3-day dosing, significantly longer terminal HL and, particularly, MRT were found (see Results; independent samples t-test).

^cPatients #1, #3, #5, #6, #9, #10 and #12 had blood group O, while the remaining patients had blood group A.

^dPearson correlations; statistically significant values (two-tailed p<0.05) are reported.

Ag: antigen; AUC: area under the curve from the time of FVIII infusion; rFVIII-Fc: recombinant FVIII-Fc fusion protein; IVR: incremental in vivo recovery; HL: terminal half-life; MRT: mean residence time; SD: standard deviation; VWF: von Willebrand Factor.

Prophylaxis regimens and patients’ follow up and outcomes on rFVIII-Fc

Duration of treatment with rFVIII-Fc (range 9–28 months) and exposure days (ED) are reported in Table III, together with changes of prophylaxis regimens over the follow up. Only 2 of 13 patients maintained their first prescribed regimen unchanged, while the others prolonged rFVIII-Fc dosing interval (Table II, panel B), ≥4 days in 9 cases. Seven of 8 patients on twice weekly regimens prolonged their dosing interval to every 4 days over 1–10 months, and one of them prolonged dosing interval further to every 5 days (#7, Table I). Among 4 patients who initially received rFVIII-Fc every 3 days, the infusion frequency was reduced twice weekly and then every 4 days over 3 months in a physically active adolescent (#5), while 2 patients (#9 and #13) were able to receive a twice weekly regimen, maintaining satisfactory trough levels with a slight dose increase. In the patient first prescribed with dosing every 4 days (#11), the interval was prolonged to every 5 days one month later, and to every 6 days after a further 6 months. No patient reverted to the previous SHL rFVIII, while 2 patients temporarily reduced their dosing interval (every 3 days, from every 4 days and twice weekly) because of intensive physiotherapy and increased physical activity, respectively. According to the last rFVIII-Fc prescribed regimen, the reduction in prophylactic infusion burden ranged from 26% to 61% (mean 40.4%, 68 infusions per year) (Table III).

Over the follow-up period, mean rFVIII-Fc weekly dose significantly decreased compared to the starting regimen (p=0.001) (Table III), in parallel with the reduced infusion frequency in most patients. Based on the last prescribed regimen, mean infusion dosing was higher on rFVIII-Fc compared with the pre-switch SHL rFVIII (46.8±4.6 vs 30.8±5.9 IU/kg; p=0.01), but mean weekly dose became statistically lower (p=0.045) (Table III). Consistently, the calculated annual rFVIII concentrate consumption was significantly reduced (mean −12.5%; p=0.019) (Table III).

Weekly rFVIII-Fc dose and individual plasma VWF:Ag showed an inverse correlation, which increased its statistical significance over treatment, when the last prescribed dose was compared to the first regimen (r= −0.66, p=0.015 and r= −0.80, p=0.001); the same finding was revealed in the inverse correlations between weekly rFVIII-Fc dose and MRT (r= −0.81, p=0.001 and r= −0.911, p <0.001) and AUC ( r= −0.703, p=0.007 and r= −0.866, p<0.001) or the direct relationships with clearance (r=0.675, p=0.011 and r=0.827, p<0.001) (Online Supplementary Content, Figure S2). Patients able to prolong the dosing interval to ≥4 days (n=9), compared with those on every 3 days or twice weekly regimens, had significantly longer terminal HL (23.5±5.9 vs 15.6±2.3 h, p=0.02) and, in particular, MRT (23.9±4.1 vs 15.7±1.3 h; p=0.005), which was >19 h in all cases (Table IV). In the same comparison, VWF:Ag levels were higher, but not statistically different (131.1±58.5 vs 69.0±6.8%; p=0.085), with 7 of 9 patients with prolonged dosing interval showing VWF:Ag levels >100%.

During rFVIII-Fc treatment on unchanged infusion interval and dosing, a trend for an increase in trough levels was observed, but differences were not statistically significant (3.1±1.9%, 3.3±1.5% and 3.8±1.9%, at first measurement, 3 and 6 months later, respectively; p>0.3). However, mean FVIII trough levels on the last rFVIII-Fc regimen were comparable to those on preswitch prophylaxis, as were bleeding rates (total and spontaneous events) (Table III); higher rates of patients did not experience any bleeding (46 vs 38%) or joint bleeds (85 vs 69%) and none had bleeds into target joints (Online Supplementary Content, Table SI); moreover, mean total FVIII (4,950±2,881 vs 9,323±6,818 IU; p=0.03), number of infusions (1.5±1.0 vs 4.1±3.4; p=0.009) and days of treatment per bleed (1.5±1.0 vs 3.6±2.5; p=0.005) were significantly lower on rFVIII-Fc than on SHL rFVIII, most events being treated with a single infusion after which prophylaxis was resumed (Online Supplementary Content, Table SI).

Adherence to the prescribed rFVIII-Fc regimens was ≥0.94 in all patients, including 3 subjects (among which the 2 eldest) showing values <0.8 on previous SHL rFVIII prophylaxis because of their taking the initiative themselves to reduce infusion frequency (from 3 to 2 per week).

Treatment satisfaction was significantly improved on rFVIII-Fc prophylaxis compared to pre-switch SHL rFVIII treatment, as revealed by the Hemo-sat scores in all domains mainly related to the pharmacological therapy (ease and convenience, efficacy, burden, general satisfaction), as well in the total score (Table V).

Table V.

Satisfaction with treatment on standard half-life (SHL) rFVIII and rFVIII-Fc prophylaxis in 13 patients, evaluated by the Hemo-sat questionnaire^a

Hemo-sat dimensionb	SHL rFVIII		rFVIII-Fc		pd
	Scoreb mean (1 SD)	Best score, nc	Scoreb mean (1 SD)	Best score, nc
Ease and convenience	27.7 (12.8)	0	19.2 (11.9)	2	0.006
Efficacy	24.7 (13.5)	0	11.8 (12.3)	4	0.003
Burden	18.3 (16.8)	3	9.8 (10.4)	5	0.005
Specialist/nurse	5.8 (11.3)	7	3.7 (5.3)	7	0.397
Centre/hospital	3.9 (7.7)	9	1.5 (4.3)	11	0.165
General satisfaction	21.2 (12.9)	2	2.9 (7.5)	11	0.001
Total score	17.7 (9.2)	0	9.7 (6.5)	1	<0.001

^aHemo-sat questionnaires were obtained on SHL rFVIII prophylaxis before switch to rFVIII-Fc and at least 6 months after starting regular prophylaxis with the latter.

^bThe questionnaire is composed of 6 dimensions with 34 items in total. Each item is scored (1–5), higher values reflecting a lower patient satisfaction; summing scores from all items yields raw scores per each domain, that are transformed in percent (0–100), considering the lowest and the highest possible values in each domain, as well in the total score. In all cases, the lower the score, the higher the level of satisfaction to treatment.

^cNumber of patients scoring 0 in each dimension.

^dComparison of scores, paired t-test.

SD: standard deviation; rFVIII: recombinant Factor VIII; rFVIII-Fc: recombinant FVIII-Fc fusion protein; SHL: standard half-life.

No adverse events clearly related to rFVIII-Fc were documented. Moreover, no inhibitor development was registered over the follow-up period, with all patients reaching >50 ED to rFVIII-Fc (10 of 13 >100 ED).

DISCUSSION

For three years or more, efmoroctocog alfa was the first (and in some countries, including Italy, the only) EHL rFVIII product available. Extensive analyses from clinical trials and expert recommendations have been published^11,26–29 but uncertainties remain among physicians and patients, as few studies clearly document real-world practice, including rates of patients switched to this (or other EHL) product, criteria for clinical choices, patients’ follow up and the optimisation of prophylaxis regimens after switching^19–21,30. Moreover, data about individual patient outcomes are very limited²⁰. We addressed these issues by reviewing the clinical experience gained at our Centre since rFVIII-Fc market availability.

The first unanswered question is how to identify the most suitable candidates to switch to the new product. Consistent with expert suggestions^13,28, in our practice, the introduction of rFVIII-Fc was driven by the careful assessment of individual clinical needs, aimed at improving the challenging management of prophylaxis. This systematic approach considered patients’ preference and convenience, preserving (or possibly increasing) efficacy and safety of treatment. The best candidates are obviously patients with venous access problems, particularly when prophylaxis feasibility or long-term adherence are compromised. Reducing IV infusion burden is the crucial factor in deciding to switch in virtually all the study population. This is consistent with a recent patient survey documenting longer dosing interval as the most reported reason to switch, whereas improving efficacy or guaranteeing the same product safety were much less rated³¹. Along these lines, another survey to rate patient preferences for prophylactic regimens reported a 2-fold higher relative importance of infusion frequency than efficacy³². Our clinical assessment, including bleeding rates on SHL rFVIII, joint status and lifestyle (with an emphasis on physical activity) highlights the importance of patient compliance to undergo the PK study and a close post-switch follow up^13,26–28. Indeed, patients should be prepared for a phase of intensive clinical/laboratory monitoring, which is crucial in order to personalise and optimise the regimens and outcomes of rFVIII-Fc prophylaxis²⁷. Patient (and parent) involvement in clinical decisions and treatment monitoring is consolidated in the haemophilia community^22–24, but their empowerment is even more relevant in the switch phase, when they need to be trained on how to adjust treatment management and lifestyle to the new product, and to identify possible changes in efficacy and safety as early as possible^26,27. On this basis, the opportunity to switch, which was shared at routine follow-up visits in most cases after patients’ (or parents’/guardians’) specific request, was considered clinically useful and feasible in approximately 60% of our patients with severe/moderate haemophilia and indications for long-term prophylaxis. Switch to rFVIII-Fc was then completed by 43% of the whole population. Our study expands the little available data about current rates of patients using EHL rFVIII. We report figures higher than the rate of 28% of severe and moderate patients on prophylaxis at the end of 2017 in the US, where EHL products became available earlier and a large variability among US regions was documented³⁰. Similarly, a recent survey conducted by the European Association of Haemophilia and Allied Disorders showed that <10% patients switched to rFVIII-Fc in most responding centres (72%) and ≥40% only in 12%²¹.

Patients’ uncertainties about switching to rFVIII-Fc concern limited clinical benefits due to the scarcely enhanced PK properties and inhibitor risk. The latter, clearly reported in an above-mentioned patient survey³¹ and by the only patient declining the opportunity to switch in our series, is unfounded in the light of clinical trials in previously treated patients with FVIII products, including EHL concentrates, and studies on national cohorts who have undergone product switch, that do not support increased inhibitor risk¹⁸. An issue of greater debate is the risk of patients with previous inhibitors who are excluded from clinical trials, in the lack of data about switching. In our series, 3 patients with previous positive inhibitor switched to rFVIII-Fc. Expert opinions suggest that patients with inhibitors eradicated by ITI within the last year should not switch FVIII concentrate^13,26 and to maintain a cautious approach in those with persistent impaired PK. However, we currently exclude patients with demanding inhibitor eradication, irrespective of timing. Another patient with long-lasting ITI-eradicated inhibitor was a candidate for switching, but his unfavourable rFVIII-Fc PK, although within the expected variability, led to rFVIII in use being maintained.

As far as clinical outcomes are concerned, all patients switching to rFVIII-Fc in our series achieved a reduction in prophylaxis infusion burden (mean −30% on first regimen), including those on highly demanding schedules (e.o.d.). Moreover, prophylaxis became feasible in a child, and adherence improved in 3 patients. As reported in literature¹⁷, and to a greater extent, over the follow-up period clinical benefits increased: 11 of 13 (85%) patients further prolonged dosing interval, ≥4 days in 9 (69%), and achieved reduction in infusions up to 62% (mean 40%). This was associated with an unchanged efficacy of treatment: FVIII trough levels and bleeding rates on rFVIII-Fc were comparable to pre-switch treatment, despite an extended dosing interval. In this respect, PK analysis and the close post-switch follow up were useful to optimise rFVIII-Fc regimens. PK-driven personalisation of prophylaxis is increasingly adopted in clinical practice, also thanks to convenient population PK tools³³; however, only HL and trough levels are considered in most cases. The influence of VWF on PK of FVIII concentrates is well recognised³⁴ and higher levels were associated with longer dosing intervals in a post-hoc analysis of the A-LONG study³⁵. Our assessments emphasise these correlations, suggesting that, more than HL itself, VWF levels and favourable PK variables, particularly MRT, may predict better outcomes over the follow-up period, with extended dosing interval and reduced rFVIII-Fc dose.

Overall, on the last rFVIII-Fc regimen, weekly dose was significantly lower than that on previous SHL rFVIII, as well the mean total annual FVIII consumption for prophylaxis. This reduction (~12%) is slightly lower than the 19% reported in preliminary Canadian analyses¹⁹ and the 17–26% from a meta-analysis and systematic literature review, which also revealed significantly lower ABR on rFVIII-Fc individualised prophylaxis than in studies with SHL rFVIII³⁶. At variance with these indirect comparisons, our study documented a real-world series of patients on highly effective pre-switch prophylaxis (mean ABR <1), most with active lifestyle and/or arthropathy. The low number of events and patients probably hampered to detect significant ABR differences compared with pre-switch treatment. However, rFVIII-Fc led to improved management of bleeds, requiring lower mean FVIII doses, number of infusions, and duration of treatment. The different patient age (most were children) and the higher pre-switch bleeding rates are likely to explain the improvement in ABR/AJBR after rFVIII-Fc switch in the only published real-world experience of 17 patients treated for mean 7 months²⁰, as well the less relevant reduction in infusion frequency (≥3 days in 35%) and the unchanged mean FVIII consumption. In this study, one patient discontinued rFVIII-Fc because of increased bruising and difficulty in performing PK evaluations²⁰. In our series, no patient reverted to SHL products and treatment satisfaction improved, as shown by the Hemo-sat scores (to our knowledge, first used to compare rFVIII-Fc and SHL prophylaxis). These findings extend data about the meaningfully improved health-related quality of life with rFVIII-Fc³⁷.

Our study may contribute to pharmaco-economic assessments about rFVIII-Fc with patient-level data, highlighting the improvement of cost-benefit ratio over treatment thanks to regimen personalisation. A budget-impact analysis from the perspective of the Italian healthcare system over a 3-year horizon showed potential savings of ~2.5% of costs for replacement products with the introduction of rFVIII-Fc³⁸. This analysis considered an uptake-rate of rFVIII-Fc increasing from 10 to 20% over 3 years and ex-factory prices for each FVIII product. The impact of FVIII acquisition costs is crucial, as shown by US data about the decrease in median annual consumption of FVIII (17%), which is not offset by the higher unit cost of EHL compared to SHL rFVIII³⁰. Inconvenient pricing strategies are likely to contribute to the relatively poor/slow uptake in many markets around the world, with reluctance to switch from SHL to rFVIII-Fc, which is still, in most cases, without competitors in its class²⁹.

CONCLUSIONS

This study provides real-world evidence about safety and efficacy of switching to rFVIII-Fc, with detailed individual patient data, including PK assessments, clinical outcomes, treatment satisfaction and costs, from a well-characterised population of all ages, followed for a meaningful observation period. This study is limited by the retrospective design and the relatively small patient population. This is, however, balanced by the accuracy and reliability of outcome data and the homogeneous clinical practice. Information about the identification of candidates to switch, based on careful assessment of patients’ needs and PK testing, and their demanding monitoring are clearly reported. Such extensive clinical work is crucial in order to personalise rFVIII-Fc regimens, thus optimising individual convenience, efficacy profiles and costs of prophylaxis. These real-world data can be useful for clinical choices in the rapidly evolving scenario of haemophilia treatment^39,40.