A Novel Patient Reported Outcome Instrument Assessing the Symptoms of Paroxysmal Nocturnal Hemoglobinuria

Introduction. Paroxysmal nocturnal hemoglobinuria (PNH) is a rare, acquired, life-threatening blood disease. Frequently reported symptoms of PNH are fatigue, dyspnea, hemoglobinuria, abdominal pain, and chest pain. While PNH-related symptoms can impact quality of life, there are no patient-reported outcome (PRO) measures that assess daily occurrence and severity of PNH-specific symptoms. The objective of this study was to gain a deeper understanding of the patient experience related to symptoms of PNH, and to assess the content validity of the newly developed PNH Symptom Questionnaire (PNH-SQ).

Methods. The content and design of the PNH-SQ was informed by a review of the empirical literature, review of COAs used in registrational PNH trials to assess symptoms and impacts of PNH, and through discussions with physicians who have experience treating patients with PNH. The first draft of the PNH-SQ allowed the patient to record, on a daily basis, the presence/absence of 12 symptoms in the past 24 hours (fatigue; shortness of breath; muscle weakness; headache; abdominal pain; leg/back pain; chest discomfort; sexual difficulties; difficulty sleeping; difficulty focusing; difficulty thinking clearly; difficulty swallowing), and the severity of each symptom using a 5-point Likert scale. Trained interviewers conducted telephone, video, or in-person qualitative, semi-structured 60-minute interviews with adult patients with a clinician-confirmed PNH diagnosis. The interviews consisted of two distinct parts: concept elicitation (CE), to assess whether the PNH-SQ captured all important symptoms from the perspective of the patient; and cognitive debriefing (CD), to evaluate the patient's ability to understand and respond to the PNH-SQ. During CE, participants discussed symptoms and impacts of PNH. This was followed by CD, in which participants reviewed and completed the PNH-SQ and asked about each part of the PNH-SQ including the instructions, understanding and relevance of items, recall period, and response options. Interviews were audio-recorded, transcribed, coded, and analyzed. Revisions to the PNH-SQ were considered after interviews were completed and analyzed.

Results. A total of 15 participants (mean age of 42.8 years; 53.3% women) from across the United States were interviewed; participants had a mean of 13.4 years since PNH diagnosis, and 66.7%, 13.3%, and 20.0% self-reported very mild/mild, moderate, and severe/very severe PNH, respectively. In the CE part of the interview, participants reported experiencing fluctuations in the presence and/or severity of individual PNH symptoms. The most common symptoms mentioned were fatigue (n=13/15), abdominal pain (n=9/15), and difficulty swallowing (n=7/15). Symptoms spontaneously mentioned by at least 5 participants were: muscle weakness; back pain; cognitive difficulties; and shortness of breath. Fatigue was by far considered the most bothersome symptom (n=8/15); other symptoms had ≤3 mentions as most bothersome symptom. Each symptom included in the PNH-SQ was spontaneously mentioned by two or more participants. Concept saturation, the point at which no new or relevant information is expected to emerge from additional interviews, was reached. In the CD segment of the interview, participants demonstrated understanding of all the symptoms included in the PNH-SQ (90-100%); considered the symptoms relevant (>50 - >90%); the recall period appropriate (80-100%); and the response options suitable (>80-100%). Based on interview analysis, two items were removed from the PNH-SQ: sexual difficulties, because it was of limited relevance to most participants (2/15), and was reported mostly by men; and difficulty focusing, because most participants (n=9/15) considered it interchangeable with difficulty thinking clearly. Minor modifications were also made to accommodate electronic administration of the PNH-SQ.

Conclusions. The PNH-SQ was developed according to FDA guidance on PRO measures development. Findings from this study support content validity of the PNH-SQ as an instrument to evaluate the daily presence and severity of PNH symptoms. In addition, the PNH-SQ has the potential to enable the evaluation of day-to-day fluctuations in PNH symptom presence and severity. The next step in instrument development is to conduct psychometric testing of the PNH-SQ using data from an ongoing clinical trial (NCT03946748).