AUTHORS
Manisha Balwani, MD, MS, Roy Smith, MD2, Paolo Ventura, MD3, Encarna Guillén-Navarro, MD, PhD4, Bruce Wang, MD5, Eliane Sardh, MD, PhD6, Penelope Stein, MB, BChir, MSc, PhD, FRCPath7, Desmond Murphy, PharmD8, Weiming Du, MS9, Rebecca Leaf, MD10, Manish Thapar, MD11
AFFILIATIONS
1 Icahn School of Medicine at Mount Sinai, New York, New York, United States
3 University of Modena and Reggio Emilia, Modena, Italy
4 CIBERER-ISCIII, Madrid, Spain, Sant Joan de Déu University Hospital, Barcelona, Spain, University of Murcia, Murcia, Spain
5 University of California San Francisco, San Francisco, California, United States
6 Karolinska University Hospital, Stockholm, Sweden
7 King’s College Hospital, London, United Kingdom
8 Alnylam Pharmaceuticals, Cambridge, Massachusetts, United States
9 Alnylam Pharmaceuticals, Cambridge, Massachusetts, United States
11 Temple University Hospital, Philadelphia, Pennsylvania, United States
ABSTRACT
Introduction
Acute hepatic porphyrias (AHPs) are rare disorders of hepatic heme synthesis characterized by potentially life-threatening acute attacks, chronic symptoms, and long-term complications, leading to poor physical and mental health, and low health-related quality of life (HRQoL). During AHP attacks, levels of heme precursors (δ-aminolevulinic acid [ALA] and porphobilinogen [PBG]) become elevated. Intravenous (IV) hemin is used to treat attacks and may be used off-label for attack prevention, but the International Porphyria Network (IPNET) guidelines recommend that this should be monitored because of risks of iron overload, thrombosis, and loss of efficacy. Givosiran, an RNA interference therapeutic approved for AHP that reduces ALA synthase production and lowers ALA/PBG levels, is the preferred treatment for attack prevention in the IPNET guidelines.
Although elevated ALA and PBG are thought to be central to disease pathology, levels observed during attacks may not fully capture disease activity. It is also unclear whether between-attack variability in ALA and PBG may influence clinical outcomes or response to givosiran treatment. This post hoc analysis of the phase 3 ENVISION study (NCT03338816) explored whether variability in ALA and PBG affected hemin use, AAR, and HRQoL following givosiran treatment.
Methods
Patients with AHP ≥12 years old with ≥2 attacks requiring hospitalization, urgent care, or IV hemin at home in the 6 months (M) before randomization entered a placebo (PBO)-controlled double-blind (DB) period followed by a 30M open-label extension (OLE), in which all patients received givosiran (givo). Patients agreed to discontinue prophylactic hemin during the study. For this post hoc analysis, patients within each treatment sequence arm (PBO/givo and givo/givo) were split into tertiles (T) by standard deviation (SD) in ALA or PBG levels recorded between 0.5M and 6M of the DB period.
Results
The analysis included 94 patients with a mean (SD) age of 39 (11) years, of whom 89% were female. In the PBO/givo and givo/givo groups, respectively, ALA tertiles were: T1, 0-6.361 mmol/mol, T2, >6.361-11.517 mmol/mol, and T3, >11.517 mmol/mol; and T1, 0-0.393 mmol/mol, T2, >0.393-1.773 mmol/mol, and T3, >1.773 mmol/mol. In the PBO/givo and givo/givo groups, respectively, PBG tertiles were: T1, 0-11.093 mmol/mol, T2, >11.093-23.323 mmol/mol, and T3, >23.323 mmol/mol; and T1, 0-1.285 mmol/mol, T2, >1.285-7.891 mmol/mol, and T3, >7.891 mmol/mol. Across both treatment arms, the proportion of patients with prior hemin prophylaxis at baseline was lower among patients with low ALA/PBG fluctuation (T1; range 25-31% of patients) than those with medium (T2; range 41-60% of patients) and high (T3; range 33-60% of patients) ALA/PBG fluctuation. Mean baseline AAR was also lower among ALA/PBG T1 than ALA/PBG T2 and T3 for both treatment arms.
With givosiran, the proportion of patients with no hemin use for attacks increased, with the highest proportions observed in the low fluctuation tertiles (T1). In the DB period, 33-69% of patients in the givo/givo group had no hemin use at >0-6M, with the same proportions for corresponding ALA and PBG tertiles. By the end of the OLE (>30-36M), this had increased to 77-100%. With givosiran initiation in the OLE (PBO/givo), the proportion of patients with no hemin use was 47-75% across ALA tertiles and 47-69% across PBG tertiles at >6-12M, increasing to 77-85% and 73-91% by >30-36M for ALA and PBG tertiles, respectively.
In most treatment arms and tertiles, a reduction in AAR from baseline was observed in the DB period that continued throughout the OLE. By >30-36M, AAR reductions exceeded 89% in every tertile, with the greatest reductions approaching 100%.
All groups reported an increase in EQ VAS scores, indicating improved HRQoL, at >30-36M, and most groups achieved the clinically meaningful increase of ≥7-8 points.
Conclusions
Before givosiran treatment, patients with higher ALA and PBG fluctuations had greater hemin use and AHP attack burden, demonstrating the clinical relevance of high levels of heme precursors in AHP. With continued givosiran treatment, more patients became hemin-free; this was concomitant with a decrease in AAR that converged across all tertiles by 36M. Improvements in HRQoL were also observed. The results support the long-term benefit of givosiran in reducing attacks, hemin use, and disease burden, which extends to patients with elevated fluctuations in AHP metabolites.
DISCLOSURES
Roy Smith, MD: No relevant conflicts of interest to declare.