Abstract
Introduction:
Hereditary hemorrhagic telangiectasia (HHT) is the second most common inherited bleeding disorder worldwide and results in substantial medical and psychosocial morbidity due to recurrent, often severe epistaxis, chronic gastrointestinal bleeding, hematologic support-dependent iron deficiency anemia, and serious bleeding and non-bleeding complications of visceral arteriovenous malformations. With disease-modifying therapeutics entering clinical trials, disease-specific health-related quality of life (HRQoL) instruments capable of longitudinal assessment of HHT-specific manifestations are essential.
Methods:
The HHT-Specific Quality of Life Scale (HHT-QoL) was developed as an easily administered, 4-item instrument assessing physical, social, personal and work impacts due to HHT longitudinally. This study successfully validates the HHT-QoL. Utilizing comprehensive, longitudinal data from PATH-HHT, the largest HHT randomized clinical trial to date, we evaluate internal consistency and external validity of the HHT-QoL and define essential characteristics for clinical application.
Results:
The HHT-QoL demonstrated good internal consistency and external validity, including directionally and clinically consistent correlations with the Epistaxis Severity Score, bleeding duration as assessed by epistaxis diary, and multiple well-validated generic HRQoL instruments measuring emotional distress, social roles and activities, and fatigue. Development of an HHT HRQoL severity classification utilizing HHT-QoL total scores was successful (0 as no limitation, 1–3 as mild, 4–7 as moderate, and 8–16 points as severe limitation) and the minimal clinically important difference (MCID) was calculated to be 1.1 points utilizing anchor-based and distribution-based methods.
Conclusions:
The HHT-QoL is a validated disease-specific HRQoL instrument with defined severity classification and MCID, making it ideal for use in clinical trials of bleeding in HHT.
Keywords: Hereditary hemorrhagic telangiectasia, Osler-Weber-Rendu, HHT, quality of life, angiogenesis, bleeding, epistaxis, gastrointestinal bleeding, anemia, iron deficiency
Graphical Abstract
INTRODUCTION
Hereditary hemorrhagic telangiectasia (HHT, Osler-Weber-Rendu disease) is an autosomal dominant vascular bleeding disorder afflicting 1 in 5,000 people, making it the second-most-common inherited bleeding disorder [1]. HHT causes recurrent, often severe epistaxis and chronic gastrointestinal bleeding from mucocutaneous telangiectases, iron deficiency anemia that may be dependent on intravenous iron and/or red cell transfusion, and arteriovenous malformations (AVMs) in the central nervous system and visceral organs causing numerous morbid and potentially fatal bleeding and non-bleeding complications [2, 3]. HHT is associated with reduced overall survival and may be the most morbid inherited bleeding disorder of women [4–8]. Bleeding in HHT, which patients and caregivers highlight as the most important disease manifestation, results in major psychosocial morbidity including social isolation, depression, anxiety, post-traumatic stress disorder, and ultimately worsened health-related quality of life (HRQoL) [9–11].
Treatment of bleeding in HHT has undergone a major paradigm shift in recent years, transitioning from a mostly procedural/surgical paradigm [12] to one primarily of effective hemostatic and antiangiogenic medical therapeutics [13–17]. While no therapeutics are yet approved for HHT by regulatory agencies, recent years have brought the first large, multicenter, randomized, placebo-controlled trials in HHT and the beginning of pharmaceutical industry investment in HHT therapeutics [18, 19]. As a result of this maturation in therapeutic development, a major unmet need for validated, disease-specific instruments to measure important clinical outcomes in HHT has emerged. Among the most critical of such needed instruments is a validated, disease-specific HRQoL measure.
HRQoL is the perception of an individual or group of their mental and physical health over time [20]. Measurement of HRQoL in clinical trials provides investigators, patients, and regulators a holistic, patient-centered measure of the risks and benefits of a given investigational therapy, integrating treatment efficacy, treatment toxicity, and the patient experience. HRQoL has repeatedly been demonstrated to be of major importance to people with HHT, particularly given the often profound impact of HHT-associated epistaxis on employment, school, social interactions, and intimate relationships [10, 12, 21, 22]. Prior work has shown that recurrent epistaxis and the consequences of iron deficiency and iron deficiency anemia (e.g., fatigue), are the clinical manifestations of HHT with the greatest impact on HRQoL. A number of well-validated generic/disease-agnostic HRQoL measurement scales are available, such as the Euro Quality of Life 5 Dimension scale (EuroQOL-5D), Short Form-36 (SF-36) and some Patient-Reported Outcomes Measurement Information System (PROMIS) measures. While these generic scales have previously demonstrated that people with HHT have reduced HRQoL compared with the general population, highlighting in particular higher burdens of depression, anxiety, discomfort, and pain in people with HHT, they have often performed poorly in HHT clinical trials, displaying minimal correlation with improvements in bleeding measures [16, 23–25]. This is likely due to the lack of items specific to epistaxis, a major manifestation of HHT impacting HRQoL. Given this, disease-specific HRQoL measures are essential in HHT.
The Hereditary Hemorrhagic Telangiectasia-Specific Quality of Life Scale (HHT-QoL) is a disease-specific HRQoL instrument comprised of four HHT-specific questions, with individual item scores summed to obtain a total score ranging from 0 to 16 (Figure 1). It was designed to be easily administered and to focus on HHT-specific impact on productivity, social interactions, and personal interactions inadequately assessed by generic instruments [22]. It was developed by a working group of HHT physician experts, patients with HHT, and methodologists and statisticians with expertise in HRQoL instruments. In a cross-sectional survey of 290 individuals with HHT, the first draft version of the HHT-QoL was demonstrated to be reliable (Cronbach’s α, 0.83) and correlate well with three well-validated PROMIS questionnaires (Discretionary Social Activities, Social Roles, and Emotional Distress) [22]. The final version was first utilized in a clinical trial as a key secondary outcome of the NIH-sponsored multicenter, randomized, double-blind, placebo-controlled PATH-HHT study of pomalidomide versus placebo to treat bleeding in HHT, where substantial improvements in the score were demonstrated with pomalidomide treatment compared with placebo [16].
Figure 1. HHT-QoL items.
A total score is calculated by summing the items 1–4 (each ranging from 0–4). The total score ranges from 0 (no limitations) to 16 (worst limitations). If any items are not answered, the total score will not be calculated. Reproduced with permission from Kasthuri et al. Blood Advances 2022 [22].
As PATH-HHT was the largest clinical trial of a therapeutic to treat HHT to date, the inclusion of the HHT-QoL as an outcome along with other efficacy measures including multiple epistaxis measures and other validated patient-reported outcomes (PROs) presents an opportunity for rigorous validation of the HHT-QoL and to better define its clinical application characteristics, such as HHT HRQoL severity classification and the minimal clinically important difference (MCID) of the instrument to optimize its use in future studies. This is the aim of the present study.
METHODS
PATH-HHT Study Assessments Relevant for HHT-QoL Validation
The PATH-HHT study enrolled 144 participants with moderate-to-severe HHT-associated bleeding, randomizing participants 2:1 to receive oral pomalidomide at a dose of 4 mg daily or placebo for 24 weeks; comprehensive methodology and results from this study, as well as the full protocol, are described in detail elsewhere [16]. Briefly, all QoL instruments, including the HHT-QoL and 3 NIH instruments (Neuro-QOL Satisfaction with Social Roles and Activities Short Form V1.1, PROMIS Emotional Distress-Depression Short Form V1.0 and PROMIS Fatigue Short Form V1.0) were administered to subjects at the baseline visit prior to study drug initiation and at the week 12 visit, week 24 visit (primary timepoint), and at the 4-week post-treatment completion follow-up visit. The Epistaxis Severity Score (ESS), a validated 10-point HHT-specific bleeding scale (the primary outcome of the study) was administered at baseline, every 4 weeks during dosing, and at the 4-week follow-up visit. A daily epistaxis diary (either electronic via a phone app or on paper) was collected over three 4-week periods: during screening prior to the baseline visit, between the week 8 and week 12 visits, and between the week 20 and week 24 visits.
All analyses are based on the intent-to-treat population from PATH-HHT (N=141 randomized, treated, and with some post-baseline efficacy data) with complete data for the HHT-QoL score at baseline (N=138), and change from baseline to 24 weeks (N=99).
Missing Data
Missing data were not imputed. If a participant did not answer an individual item, the data were considered missing for that item, and the total score was not calculated. Missing data for total scores were handled with listwise deletion in analyses.
Internal Consistency, Construct Validity, and Determination of Severity Classification Thresholds
The frequency distribution of responses for each HHT-QoL item and the HHT-QoL total score at the baseline assessment were evaluated. The reliability (internal consistency) of the HHT-QoL items was assessed utilizing Cronbach’s α and Spearman correlations between each of the four HHT-QoL items with each of the other three items and the HHT-QoL total score (both for the baseline assessment as well as the change in score from baseline to 24 weeks). Spearman correlations were used for all analyses to minimize impact of any outliers. Construct validity was addressed by principal components analysis of the HHT-QoL items. The frequency distribution of the total scores at baseline was evaluated to determine optimal thresholds for the definitions of mild impairment, moderate impairment, and severe impairment in HRQoL due to HHT-specific disease manifestations. Following the determination of optimal thresholds, participants were grouped by severity classification and the suitability and consistency of the classification was assessed via descriptive statistics of responses to each HHT-QoL item by severity class at baseline.
External Validity with Epistaxis Instruments and Generic HRQoL Instruments
To evaluate external validity and association of the HHT-QoL instrument with efficacy assessments and other validated HRQoL instruments (ESS, daily bleeding duration as assessed by epistaxis diary, Neuro-QOL Satisfaction with Social Roles and Activities Short Form V1.1, PROMIS Emotional Distress-Depression Short Form V1.0 and PROMIS Fatigue Short Form V1.0), Spearman correlations of the HHT-QoL items and total score with the total score of each other assessment were calculated for both the baseline assessment as well as the change from baseline to 24 weeks.
Determination of the Minimal Clinically Important Difference of the HHT-QoL
The MCID of the HHT-QoL was determined using both anchor- and distribution-based methods and then averaging together the MCID estimates for each anchor and the distribution-based method [26–29]. Suitable anchors for the anchor-based method were required to have a correlation with HHT QoL for change from baseline of at least r ≥ 0.3 or ≤ −0.3 [28, 29]. For each anchor, the MCID was computed as the difference in mean change scores for those reporting a small (but clinically relevant) improvement vs. no change. For the ESS and NeuroQol Satisfaction with Social Roles and Activities, the MCID or conditional minimal detectable change [30] was used as the threshold for small improvement, and for the bleeding duration by daily epistaxis diary a threshold of ≥5 minutes per day (equating to ≥140 minutes over the 4-week evaluation period, an objectively robust improvement) was used as the threshold for small improvement.
Distribution-based MCID estimates were based on a percentage of the baseline standard deviation (SD), where 0.5 SD is considered a medium effect size and 0.2 SD is considered a small effect size [26, 29]. To ensure a rigorous estimate was used, the MCID estimate based on a 0.5 SD (medium effect size) was utilized in the overall MCID calculation.
RESULTS
Internal Consistency and Construct Validity
The frequency distribution of responses at baseline for each HHT-QoL item are illustrated in Figure 2 and the frequency distribution of responses at baseline for the HHT-QoL total score is illustrated in Figure 3. Of the 141 participants in the intent-to-treat population, the HHT QoL total score was available for 138 (98%). One participant had a missing form, one answered only the first item, and another participant did not respond to item 2. Each of the items demonstrates scores across the range of responses from “never” to “always”, with items 1 and 2 approximating a normal distribution and a right skew for item 4. The total score, which is calculated as a sum of the 4 items, approximated a normal distribution.
Figure 2. Distribution of HHT-QoL item answers at baseline in PATH-HHT study.
Percentages on the Y-axis represent the percent of participants in the PATH-HHT study ITT population with non-missing HHT-QoL total scores (n=138).
Figure 3. Distribution of HHT-QoL total score at baseline in PATH-HHT study, with HHT-specific HRQoL severity classification.
Percentages on the Y-axis represent the percent of participants in the ITT population with non-missing HHT-QoL total scores (n=138).
The HHT-QoL demonstrated good internal consistency. HHT-QoL items 1–3, all of which measure impact of epistaxis, correlated moderately with each other at baseline (pairwise correlations of 0.42–0.74, Table 1) and strongly with HHT-QoL total score at baseline (pairwise correlations of 0.76–0.84). Similarly strong pairwise correlations were observed in the change in scores from baseline to 24 weeks (Table 2). Item 4, which measures impact of all non-epistaxis HHT manifestations, correlated weakly with the other items and moderately with the total score (Table 1 and Table 2). The Cronbach’s α for all 4 items was 0.74, confirming that the items form a reliable scale. The Cronbach’s α with one of the 4 items removed were 0.63 (item 1 removed), 0.60 (item 2 removed), 0.69 (item 3 removed), and 0.78 (item 4 removed), indicating that although item 4 was less highly correlated with the other 3 items, Cronbach’s α with item 4 removed was similar to the score with all 4 items included.
Table 1.
Spearman correlations of baseline HHT-QoL, epistaxis measures, and validated generic HRQoL instruments.
| HHT-QoL Item 1 | HHT-QoL Item 2 | HHT-QoL Item 3 | HHT-QoL Item 4 | HHT-QoL Total score | |
|---|---|---|---|---|---|
|
| |||||
| HHT-QoL Item 1 | 0.74 | 0.42 | 0.23 | 0.79 | |
| HHT-QoL Item 2 | 0.49 | 0.27 | 0.84 | ||
| HHT-QoL Item 3 | 0.29 | 0.76 | |||
| HHT-QoL Item 4 | 0.56 | ||||
|
| |||||
| ESS Total Score | 0.38 | 0.26 | 0.26 | 0.30 | 0.38 |
| Bleeding duration diary | 0.43 | 0.28 | 0.22 | 0.22 | 0.35 |
| Weighted bleeding duration - diary | 0.45 | 0.33 | 0.24 | 0.24 | 0.38 |
| Neuro-QoL Satisfaction with Social Roles and Activities | −0.30 | −0.30 | −0.48 | −0.41 | −0.49 |
| PROMIS Emotional Distress - Depression | 0.13 | 0.17 | 0.34 | 0.37 | 0.32 |
| PROMIS Fatigue | 0.42 | 0.39 | 0.25 | 0.49 | 0.49 |
Intent-to-treat population with non-missing HHT-QoL total score, N=138 (N=137 for Neuro-QoL, n=129 for bleeding diary).
The HHT-QoL score ranges from 0 to 16 with higher scores indicating more limitations due to HHT in the prior 4 weeks.
The daily epistaxis duration is calculated as the total duration of all reported nose bleeding events in a day, averaged across all days reported in a 4-week smartphone diary. Reporting no events in a day is given a 0 duration.
ESS is the Epistaxis Severity Score.
The weighted daily epistaxis duration is calculated as the total duration of all reported nose bleeding events in a day, averaged across all days reported in a 4-week smartphone diary, with each duration weighted by the intensity as follows: spotting: 0.5, dripping: 1, dripping quickly: 1.5, steady stream: 2, pouring: 2.5, gushing: 3.
The Neuro-QoL™ Satisfaction with Social Roles and Activities Short Form (V1.1) T-score has a value of 50 representing the average general US population, with a standard deviation of 10, and with higher scores indicating more satisfaction.
The PROMIS® Emotional Distress-Depression Short Form (V1.0) T-score has a value of 50 representing the average general US population, with a standard deviation of 10, and with higher scores indicating more depression.
The PROMIS® Fatigue Short Form (V1.0) T-score has a value of 50 representing the average general US population, with a standard deviation of 10, and with higher scores indicating more fatigue.
Correlation shading categories: ≥0.25 - <0.4 (weak), 0.4 - <0.75 (moderate), ≥0.75 (strong).
Table 2.
Spearman correlations of change in scores from baseline to 24 weeks for HHT-QoL, epistaxis measures, and validated generic HRQoL instruments.
| Change from baseline | HHT-QoL Item 1 | HHT-QoL Item 2 | HHT-QoL Item 3 | HHT-QoL Item 4 | HHT-QoL Total score |
|---|---|---|---|---|---|
|
| |||||
| HHT-QoL Item 1 | 0.64 | 0.44 | 0.13 | 0.78 | |
| HHT-QoL Item 2 | 0.48 | 0.11 | 0.82 | ||
| HHT-QoL Item 3 | 0.05 | 0.69 | |||
| HHT-QoL Item 4 | 0.43 | ||||
|
| |||||
| ESS Total Score | 0.35 | 0.39 | 0.33 | 0.13 | 0.44 |
| Bleeding duration diary | 0.47 | 0.37 | 0.18 | 0.13 | 0.38 |
| Weighted bleeding duration - diary | 0.47 | 0.39 | 0.18 | 0.16 | 0.39 |
| Neuro-QoL Satisfaction with Social Roles and Activities | −0.30 | −0.28 | −0.34 | −0.22 | −0.38 |
| PROMIS Emotional Distress - Depression | 0.14 | 0.11 | 0.24 | 0.28 | 0.24 |
| PROMIS Fatigue | 0.22 | 0.14 | 0.20 | 0.14 | 0.27 |
Intent-to-treat population with non-missing HHT-QoL total score change from baseline at 24 weeks, N=99 (N=98 for Neuro-QoL, n=79 for bleeding diary)
The HHT-QoL score ranges from 0 to 16 with higher scores indicating more limitations due to HHT in the prior 4 weeks.
The daily epistaxis duration is calculated as the total duration of all reported nose bleeding events in a day, averaged across all days reported in a 4-week smartphone diary. Reporting no events in a day is given a 0 duration.
ESS is the Epistaxis Severity Score.
The weighted daily epistaxis duration is calculated as the total duration of all reported nose bleeding events in a day, averaged across all days reported in a 4-week smartphone diary, with each duration weighted by the intensity as follows: spotting: 0.5, dripping: 1, dripping quickly: 1.5, steady stream: 2, pouring: 2.5, gushing: 3.
The Neuro-QoL™ Satisfaction with Social Roles and Activities Short Form (V1.1) T-score has a value of 50 representing the average general US population, with a standard deviation of 10, and with higher scores indicating more satisfaction.
The PROMIS® Emotional Distress-Depression Short Form (V1.0) T-score has a value of 50 representing the average general US population, with a standard deviation of 10, and with higher scores indicating more depression.
The PROMIS® Fatigue Short Form (V1.0) T-score has a value of 50 representing the average general US population, with a standard deviation of 10, and with higher scores indicating more fatigue.
Correlation shading categories: ≥0.25 - <0.4 (weak), 0.4 - <0.75 (moderate), ≥0.75 (strong).
The principal components analysis (Figure 4) confirmed that the scale was unidimensional. The first principal component had an Eigenvalue of 2.3 and explained 58% of the variance, and all other principal components had Eigenvalues <1 and added minimal variance explanation. This confirmed that a single-component solution is appropriate.
Figure 4. Principal components analysis of HHT-QoL items at baseline.
Determination of Severity Classification Thresholds
On the basis of the frequency distribution for the total score, severity classification thresholds were determined as follows: 0 = no impairment, 1–3 = mild impairment, 4–7 = moderate impairment, and 8–16 = severe impairment due to HHT during the preceding 4 weeks In this study population of participants with moderate to severe HHT-associated bleeding, less than 2% had no impairment and 18% reported mild impairment. An evaluation of HHT-QoL item result descriptive statistics at baseline by severity classification (Table 3) demonstrated consistent, appropriate stepwise increases in score for each item across all severity classes, confirming the suitability of the severity classification thresholds.
Table 3.
Descriptive statistics of the HHT-QoL at baseline by severity classification.
| Mean (SD) Median [IQR] | HHT-QoL Item 1 | HHT-QoL Item 2 | HHT-QoL Item 3 | HHT-QoL Item 4 | HHT-QoL Total score |
|---|---|---|---|---|---|
|
| |||||
| All (n=138) | 2.2 (1.0) | 1.9 (1.0) | 1.4 (1.1) | 0.8 (1.0) | 6.2 (3.1) |
| 2 [2, 3] | 2 [1, 3] | 1 [0, 2] | 0 [0, 1] | 4 [6, 8] | |
| HHT-QoL 1–3 (mild, n=25) | 0.9 (0.6) | 0.8 (0.6) | 0.3 (0.6) | 0.2 (0.6) | 2.2 (0.8) |
| 1 [1, 1] | 1 [0, 1] | 0 [0, 1] | 0 [0, 0] | 2 [2, 3] | |
| HHT-QoL 4–7 (moderate, n=66) | 2.2 (0.7) | 1.8 (0.6) | 1.1 (0.8) | 0.5 (0.7) | 5.6 (1.0) |
| 2 [2, 3] | 2 [1, 2] | 1 [0, 2] | 0 [0, 1] | 6 [5, 6] | |
| HHT-QoL 8–16 (severe, n=45) | 3.0 (0.5) | 2.8 (0.6) | 2.4 (0.9) | 1.5 (1.1) | 9.8 (1.7) |
| 3 [3, 3] | 3 [2, 3] | 2 [2, 3] | 1 [1, 2] | 9 [9, 11] | |
Intent-to-treat population with non-missing HHT-QoL total score, N=138.
The HHT-QoL score ranges from 0 to 16 with higher scores indicating more limitations due to HHT in the prior 4 weeks.
Correlations with Epistaxis Instruments and Generic HRQoL Instruments
The HHT-QoL instrument also demonstrated good external validity. The HHT-QoL total score demonstrated weakly to moderately strong, directionally and clinically consistent correlations with the ESS, daily bleeding duration as assessed by epistaxis diary (both intensity-weighted and without weighting), Neuro-QOL Satisfaction with Social Roles and Activities Short Form V1.1, PROMIS Emotional Distress-Depression Short Form V1.0 and PROMIS Fatigue Short Form V1.0 at baseline (Table 1). Similarly strong and directionally consistent correlations were observed with change in scores from baseline to 24 weeks between the HHT-QoL total score and all of these efficacy measures except for PROMIS Emotional Distress-Depression Short Form V1.0 (Table 2). In comparison, each of the three generic HRQoL measures were not correlated with the ESS or bleeding diary at baseline or change from baseline at 24 weeks (all Spearman correlations ≤ 0.2).
Determination of the Minimal Clinically Important Difference of the HHT-QoL
Based on the anchor eligibility criterion (Spearman correlation coefficients for change from baseline r ≥ 0.3 or ≤ −0.3), three anchors were utilized in the anchor-based MCID analysis: ESS, intensity-weighted daily bleeding duration as assessed by epistaxis diary, and Neuro-QOL Satisfaction with Social Roles and Activities Short Form V1.1. While both the weighted and unweighted form of bleeding duration met the correlation criteria, we selected the weighted measure for MCID calculation as it encompasses more information on epistaxis (Table 4). The MCID estimates from each of these anchors, respectively, were 1.1, 0.5, and 1.4 points. The baseline standard deviation of the HHT-QoL total score was 3.08, with 0.2 SD (small effect size) therefore equaling 0.6 points and 0.5 SD (medium effect size) equaling 1.5 points.
Table 4.
Anchor-based MCID estimates for the HHT-QoL.
| HHT-QoL Change from Baseline to 24 weeks | |||
|---|---|---|---|
|
| |||
| Criterion | N | Mean (SD) | MCID estimatea |
|
| |||
| ESS Total Score change at 24 weeks | |||
| Large Improvement (< −1.42) | 48 | −3.0 (2.8) | |
| Small improvement (> −1.42 to −0.71) | 18 | −1.9 (3.7) | |
| No improvement (> −0.71 to < 0.71) | 14 | −0.8 (3.0) | (−1.9 – −0.8) = 1.1 |
| Decline (>=0.71) | 19 | −0.3 (2.7) | |
| Bleeding Diary Severity-weighted Minutes per day change at 24 weeks | |||
| Large Improvement (< −10) | 44 | −2.5 (3.1) | |
| Small improvement (> −10 to −5) | 18 | −2.6 (2.6) | (−2.6 – −2.1) = 0.5 |
| No improvement (> −5 to < 5) | 24 | −2.1(3.0) | |
| Decline (>= 5) | 13 | 0.9 (3.3) | |
| NeuroQol Satisfaction with Social Roles T score change at 24 weeks | |||
| Large Improvement (>=8) | 12 | −4.7 (3.7) | |
| Small improvement (4 to <8) | 13 | −2.9 (1.9) | (−2.9 – −1.5) = 1.4 |
| No improvement (> −4 to <4) | 61 | −1.5 (2.7) | |
| Decline (<= −4) | 13 | −0.5 (4.2) | |
For each anchor, the MCID was computed as the difference in mean change scores for those reporting a small amount of improvement vs. no change.
ESS is the Epistaxis Severity Score
The overall MCID of the HHT-QoL was therefore calculated to be 1.1 points (or 1 point when rounded to the nearest integer) based on an average of the MCID estimates for each anchor and the MCID estimate for the 0.5 SD (medium effect size) from the distribution-based method.
DISCUSSION
In this analysis, we confirm the construct validity, internal consistency and external validity of the HHT-QoL instrument utilizing data from PATH-HHT, the largest clinical trial to date of the treatment of HHT. Importantly, we also determine the MCID of the instrument (1.1 points) and suitable severity classification thresholds for the instrument, with 0 as no impairment, 1–3 as mild impairment, 4–7 as moderate impairment, and 8–16 as severe impairment due to HHT during the preceding 4 weeks. With this, the HHT-QoL is the first disease-specific HRQoL instrument to undergo successful validation in a large randomized-controlled trial and the first with suitably developed severity classification thresholds and MCID.
While HHT is a heterogenous bleeding disorder, the clinical manifestation that unites over 95% of patients is recurrent, often severe epistaxis. While recurrent epistaxis may lead to non-anemic iron deficiency or iron deficiency anemia, thereby resulting in hemoglobin drops and hematologic support requirements with intravenous iron and/or red cell transfusion, the most pronounced hematologic consequences of HHT are driven by chronic gastrointestinal bleeding (which occurs in a minority of patients with HHT) [31]. This means that hematologic endpoints such as hemoglobin or hematologic support requirements may be inadequate as standalone measures for evaluating the impact of therapeutics on improving epistaxis. Furthermore, patients with mild or moderate epistaxis as classified by epistaxis instruments or an epistaxis diary, both of which focus on objective aspects of epistaxis such as duration and frequency, may experience more profound HRQoL impacts from supposedly “mild” or “moderate” objective measures of epistaxis. This can include employment challenges and unemployment, physical inactivity, social isolation and sexual abstention, among many others. In a recent cross-sectional study, over 40% of patients with HHT carried diagnoses of major depression, an anxiety disorder, or post-traumatic stress disorder [10], and much of this psychiatric morbidity appears to be secondary to recurrent and often unpredictable epistaxis. The ESS, epistaxis diaries, and hematologic measurements relied upon in clinical trials do not assess mental health comorbidities. They also do not assess the impact of epistaxis on work, school, intimate partnership, family, friends, social activities, or other regularly scheduled commitments, all of which are assessed by the HHT-QoL. The limitations of our epistaxis severity instruments and hematologic measures to evaluate disease burden in HHT underscore the importance of an easily administered, validated instrument to evaluate the HRQoL impact of all aspects of HHT, both epistaxis-specific and non-epistaxis specific, as can be accomplished by the HHT-QoL, which will complement the objective epistaxis and hematologic measures in both patient care and clinical research settings.
While the MCID for the HHT-QoL calculated in this study was 1.1 points, the scoring of the instrument (Figure 1) is such that all total scores will be an integer. Therefore, when using the MCID in individual patients, we recommend rounding 1.1 points down to 1 point. Of note, the between-group difference between the pomalidomide and placebo groups in the HHT-QoL in the PATH-HHT trial was −1.4 points (95% CI −2.6 to −0.3) [16], exceeding the MCID as determined in the present study. No between-group differences were noted for the 3 generic HRQoL instruments. It should be noted that MCID estimates can vary across disparate patient subgroups, such as those with mild disease versus severe disease. The calculated MCID of the HHT-QoL of 1.1 points (rounded to 1 point) in this study is therefore most applicable in future studies of patients with moderate-to-severe HHT-associated bleeding, as was the case in PATH-HHT. Similarly, it is also relevant to note that while the derivation of the instrument was performed in a broad HHT population spanning from mild to severe disease, the external validation in PATH-HHT was completed in patients with moderate-to-severe HHT-associated bleeding, making the HRQoL severity classification most applicable in this population as well.
The HHT-QoL is not the only disease-specific HRQoL instrument in HHT that has been developed, as there is one other recently described instrument, the Quality of Life Questionnaire in HHT [21]. This instrument, which to date is only available in French, was developed and validated in a cross-sectional study in a group of patients with HHT in France but has not yet undergone validation in a clinical trial or been used to assess the impact of an intervention. It includes 24 items that assess six specific domains. No severity classification or MCID is currently available for this instrument, and no studies to date have compared this instrument with the HHT-QoL. Given the inclusion in the Quality of Life Questionnaire in HHT of numerous items that are not expected to change with therapy over the course of a clinical trial (such as items assessing the patient’s worry about the health of their family members with HHT and their therapeutic relationship with their healthcare providers), it may be better suited to cross-sectional evaluation of HRQoL in HHT as opposed to the HHT-QoL, which was designed for longitudinal assessment to measure the evolution in disease-specific HRQoL over time as an outcome in HHT clinical trials.
In summary, this study validated the HHT-QoL and determined the characteristics of the instrument necessary for its wide-ranging clinical application. The brevity of the HHT-QoL, its focus on the psychosocial impacts of epistaxis, inclusion of non-epistaxis impact of HHT, and the now-defined severity classification and MCID of the instrument make it an ideal option to include as an endpoint in clinical trials of therapeutics to treat bleeding in HHT. Because the HHT-QoL was developed with funding from the National Heart, Lung, and Blood Institute of the U.S. National Institutes of Health, it is freely available for use in clinical trials and clinical practice without any required permissions or monetary cost. All of these features make it a logical choice for regulatory agencies to recognize as a preferred PRO in HHT clinical studies. As therapeutic development in HHT continues to accelerate, with the entry of multiple pharmaceutical companies into the HHT drug development space in addition to ongoing investigator-initiated clinical trials repurposing antiangiogenic cancer drugs, the HHT-QoL may be easily and routinely utilized in HHT clinical studies moving forward.
ACKNOWLEDGEMENTS
The authors would like to thank the patients with HHT who participated in the cross-sectional survey allowing for the initial development of the HHT-QoL as well as the patients who participated in the PATH-HHT clinical trial and the investigators and support staff of this trial, who helped to generate the data from which have made the present study possible.
Sources of support (financial, equipment, drugs):
This study was supported by grants (UG3HL140097 to K. McCrae, U24HL140090 to S. Thomas, and K23HL159313 to H. Al-Samkari) from the National Heart, Lung, and Blood Institute.
Footnotes
CONFLICT OF INTEREST STATEMENT
H. Al-Samkari: Grants or contracts (research funding to institution) (Agios, Sobi, Vaderis, Novartis, Amgen), consulting fees (Agios, Amgen, Alnylam, Alpine, Sobi, argenx, Pharmacosmos, Novartis, Sanofi). R. Kasthuri: Grants or contracts (research funding to institution) (U.S. Department of Defense, HRSA), consulting fees (Alnylam, Crosswalk, Pharmacosmos, Tectonic). S. Thomas: Grants or contracts (research funding to institution) (NHLBI). K. McCrae: Payment or honoraria for lectures/presentations/other (MD Anderson Cancer Center, University of Southern California), participation in data safety monitoring board/advisory board (Alnylam). A. Pishko: Advisory board (Biomarin LLC), royalties (UptoDate). All other authors report no relevant disclosures.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
REFERENCES
- 1.Kjeldsen AD, Vase P, Green A. Hereditary haemorrhagic telangiectasia: a population-based study of prevalence and mortality in Danish patients. J Intern Med. 1999; 245: 31–9. 10.1046/j.1365-2796.1999.00398.x. [DOI] [PubMed] [Google Scholar]
- 2.Kasthuri RS, Montifar M, Nelson J, Kim H, Lawton MT, Faughnan ME, Brain Vascular Malformation Consortium HHTIG. Prevalence and predictors of anemia in hereditary hemorrhagic telangiectasia. Am J Hematol. 2017. 10.1002/ajh.24832. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kjeldsen AD, Kjeldsen J. Gastrointestinal bleeding in patients with hereditary hemorrhagic telangiectasia. Am J Gastroenterol. 2000; 95: 415–8. 10.1111/j.1572-0241.2000.01792.x. [DOI] [PubMed] [Google Scholar]
- 4.Donaldson JW, McKeever TM, Hall IP, Hubbard RB, Fogarty AW. Complications and mortality in hereditary hemorrhagic telangiectasia: A population-based study. Neurology. 2015; 84: 1886–93. 10.1212/WNL.0000000000001538. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Thompson KP, Nelson J, Kim H, Pawlikowska L, Marchuk DA, Lawton MT, Faughnan ME, Brain Vascular Malformation Consortium HHTIG. Predictors of mortality in patients with hereditary hemorrhagic telangiectasia. Orphanet J Rare Dis. 2021; 16: 12. 10.1186/s13023-020-01579-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zhang E, Virk ZM, Rodriguez-Lopez J, Al-Samkari H. Hereditary hemorrhagic telangiectasia may be the most morbid inherited bleeding disorder in women. Blood advances. 2024; 8: 3166–72. 10.1182/bloodadvances.2023011961. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kritharis A, Al-Samkari H, Kuter DJ. Hereditary hemorrhagic telangiectasia: diagnosis and management from the hematologist’s perspective. Haematologica. 2018; 103: 1433–43. 10.3324/haematol.2018.193003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mora-Lujan JM, Iriarte A, Alba E, Sanchez-Corral MA, Cerda P, Cruellas F, Ordi Q, Corbella X, Ribas J, Castellote J, Riera-Mestre A. Gender differences in hereditary hemorrhagic telangiectasia severity. Orphanet J Rare Dis. 2020; 15: 63. 10.1186/s13023-020-1337-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Shovlin CL, Buscarini E, Sabba C, Mager HJ, Kjeldsen AD, Pagella F, Sure U, Ugolini S, Torring PM, Suppressa P, Rennie C, Post MC, Patel MC, Nielsen TH, Manfredi G, Lenato GM, Lefroy D, Kariholu U, Jones B, Fialla AD, Eker OF, Dupuis O, Droege F, Coote N, Boccardi E, Alsafi A, Alicante S, Dupuis-Girod S. The European Rare Disease Network for HHT Frameworks for management of hereditary haemorrhagic telangiectasia in general and speciality care. Eur J Med Genet. 2022; 65: 104370. 10.1016/j.ejmg.2021.104370. [DOI] [PubMed] [Google Scholar]
- 10.Chaturvedi S, Clancy M, Schaefer N, Oluwole O, McCrae KR. Depression and post-traumatic stress disorder in individuals with hereditary hemorrhagic telangiectasia: A cross-sectional survey. Thromb Res. 2017; 153: 14–8. 10.1016/j.thromres.2017.03.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.OS AA, Friedman CM, White RI Jr. The natural history of epistaxis in hereditary hemorrhagic telangiectasia. Laryngoscope. 1991; 101: 977–80. 10.1288/00005537-199109000-00008. [DOI] [PubMed] [Google Scholar]
- 12.Faughnan ME, Mager JJ, Hetts SW, Palda VA, Lang-Robertson K, Buscarini E, Deslandres E, Kasthuri RS, Lausman A, Poetker D, Ratjen F, Chesnutt MS, Clancy M, Whitehead KJ, Al-Samkari H, Chakinala M, Conrad M, Cortes D, Crocione C, Darling J, de Gussem E, Derksen C, Dupuis-Girod S, Foy P, Geisthoff U, Gossage JR, Hammill A, Heimdal K, Henderson K, Iyer VN, Kjeldsen AD, Komiyama M, Korenblatt K, McDonald J, McMahon J, McWilliams J, Meek ME, Mei-Zahav M, Olitsky S, Palmer S, Pantalone R, Piccirillo JF, Plahn B, Porteous MEM, Post MC, Radovanovic I, Rochon PJ, Rodriguez-Lopez J, Sabba C, Serra M, Shovlin C, Sprecher D, White AJ, Winship I, Zarrabeitia R. Second International Guidelines for the Diagnosis and Management of Hereditary Hemorrhagic Telangiectasia. Ann Intern Med. 2020; 173: 989–1001. 10.7326/M20-1443. [DOI] [PubMed] [Google Scholar]
- 13.Al-Samkari H Hereditary Hemorrhagic Telangiectasia: Systemic Therapies, Guidelines, and an Evolving Standard of Care. Blood. 2020. 10.1182/blood.2020008739. [DOI] [PubMed] [Google Scholar]
- 14.Al-Samkari H, Kasthuri RS, Parambil JG, Albitar HA, Almodallal YA, Vazquez C, Serra MM, Dupuis-Girod S, Wilsen CB, McWilliams JP, Fountain EH, Gossage JR, Weiss CR, Latif MA, Issachar A, Mei-Zahav M, Meek ME, Conrad M, Rodriguez-Lopez J, Kuter DJ, Iyer VN. An international, multicenter study of intravenous bevacizumab for bleeding in hereditary hemorrhagic telangiectasia: the InHIBIT-Bleed study. Haematologica. 2021; 106: 2161–9. 10.3324/haematol.2020.261859. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Invernizzi R, Quaglia F, Klersy C, Pagella F, Ornati F, Chu F, Matti E, Spinozzi G, Plumitallo S, Grignani P, Olivieri C, Bastia R, Bellistri F, Danesino C, Benazzo M, Balduini CL. Efficacy and safety of thalidomide for the treatment of severe recurrent epistaxis in hereditary haemorrhagic telangiectasia: results of a non-randomised, single-centre, phase 2 study. The Lancet Haematology. 2015; 2: e465–73. 10.1016/S2352-3026(15)00195-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Al-Samkari H, Kasthuri RS, Iyer VN, Pishko AM, Decker JE, Weiss CR, Whitehead KJ, Conrad MB, Zumberg MS, Zhou JY, Parambil J, Marsh D, Clancy M, Bradley L, Wisniewski L, Carper BA, Thomas SM, McCrae KR. Pomalidomide for Epistaxis in Hereditary Hemorrhagic Telangiectasia. N Engl J Med. 2024; 391: 1015–27. 10.1056/NEJMoa2312749. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Parambil JG, Gossage JR, McCrae KR, Woodard TD, Menon KVN, Timmerman KL, Pederson DP, Sprecher DL, Al-Samkari H. Pazopanib for severe bleeding and transfusion-dependent anemia in hereditary hemorrhagic telangiectasia. Angiogenesis. 2022; 25: 87–97. 10.1007/s10456-021-09807-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Al-Samkari H Giving hereditary haemorrhagic telangiectasia the attention it deserves. The Lancet Haematology. 2021; 8: e472–e4. 10.1016/S2352-3026(21)00164-2. [DOI] [PubMed] [Google Scholar]
- 19.Al-Samkari H, Hessels J, Riera-Mestre A, Dupuis-Girod S, Van Zele T, Gomez del Olmo V, Saint-Mezard P, Lazar H, Picard D, Barker D, Buscarini E, Mager HJ. A Randomized, Placebo-Controlled, Multicenter Proof-of-Concept (POC) Study to Assess the Safety and Efficacy of the Novel Allosteric AKT Inhibitor, VAD044, in Adults with Hereditary Hemorrhagic Telangiectasia (HHT). Blood. 2024; 144: 553. [Google Scholar]
- 20.Study protocol for the World Health Organization project to develop a Quality of Life assessment instrument (WHOQOL). Qual Life Res. 1993; 2: 153–9. [PubMed] [Google Scholar]
- 21.Le TTT, Martinent G, Dupuis-Girod S, Parrot A, Contis A, Riviere S, Chinet T, Grobost V, Espitia O, Dussardier-Gilbert B, Alric L, Armengol G, Maillard H, Leguy-Seguin V, Leroy S, Rondeau-Lutz M, Lavigne C, Mohamed S, Chaussavoine L, Magro P, Seguier J, Kerjouan M, Fourdrinoy S. Development and validation of a quality of life measurement scale specific to hereditary hemorrhagic telangiectasia: the QoL-HHT. Orphanet J Rare Dis. 2022; 17: 281. 10.1186/s13023-022-02426-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kasthuri RS, Chaturvedi S, Thomas S, Vandergrift N, Bann C, Schaefer N, Clancy MS, Pyeritz R, McCrae KR. Development and performance of a hereditary hemorrhagic telangiectasia-specific quality-of-life instrument. Blood advances. 2022; 6: 4301–9. 10.1182/bloodadvances.2022007748. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Pasculli G, Resta F, Guastamacchia E, Di Gennaro L, Suppressa P, Sabba C. Health-related quality of life in a rare disease: hereditary hemorrhagic telangiectasia (HHT) or Rendu-Osler-Weber disease. Qual Life Res. 2004; 13: 1715–23. 10.1007/s11136-004-7865-y. [DOI] [PubMed] [Google Scholar]
- 24.Geisthoff UW, Heckmann K, D’Amelio R, Grunewald S, Knobber D, Falkai P, Konig J. Health-related quality of life in hereditary hemorrhagic telangiectasia. Otolaryngol Head Neck Surg. 2007; 136: 726–33; discussion 34–5. 10.1016/j.otohns.2006.12.019. [DOI] [PubMed] [Google Scholar]
- 25.Hermann R, Grobost V, Le-Guillou X, Lavigne C, Parrot A, Riviere S, Seguier J, Fargeton AE, de-Montigny A, Huot M, Decullier E, Roux A, Gervaise C, Cartier C, Dufour X, Grall M, Jegoux F, Laccourreye L, Michel J, Saroul N, Wagner I, Kerjouan M, Dupuis-Girod S. Effect of oral nintedanib vs placebo on epistaxis in hereditary hemorrhagic telangiectasia: the EPICURE multicenter randomized double-blind trial. Angiogenesis. 2024; 28: 9. 10.1007/s10456-024-09962-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Crosby RD, Kolotkin RL, Williams GR. Defining clinically meaningful change in health-related quality of life. J Clin Epidemiol. 2003; 56: 395–407. 10.1016/s0895-4356(03)00044-1. [DOI] [PubMed] [Google Scholar]
- 27.Revicki D, Hays RD, Cella D, Sloan J. Recommended methods for determining responsiveness and minimally important differences for patient-reported outcomes. J Clin Epidemiol. 2008; 61: 102–9. 10.1016/j.jclinepi.2007.03.012. [DOI] [PubMed] [Google Scholar]
- 28.Guyatt GH, Osoba D, Wu AW, Wyrwich KW, Norman GR, Clinical Significance Consensus Meeting G. Methods to explain the clinical significance of health status measures. Mayo Clin Proc. 2002; 77: 371–83. 10.4065/77.4.371. [DOI] [PubMed] [Google Scholar]
- 29.Wyrwich KW, Bullinger M, Aaronson N, Hays RD, Patrick DL, Symonds T, Clinical Significance Consensus Meeting G. Estimating clinically significant differences in quality of life outcomes. Qual Life Res. 2005; 14: 285–95. 10.1007/s11136-004-0705-2. [DOI] [PubMed] [Google Scholar]
- 30.Kozlowski AJ, Cella D, Nitsch KP, Heinemann AW. Evaluating Individual Change With the Quality of Life in Neurological Disorders (Neuro-QoL) Short Forms. Arch Phys Med Rehabil. 2016; 97: 650–4 e8. 10.1016/j.apmr.2015.12.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Mora-Lujan JM, Iriarte A, Alba E, Sanchez-Corral MA, Berrozpe A, Cerda P, Cruellas F, Ribas J, Castellote J, Riera-Mestre A. Gastrointestinal Bleeding in Patients with Hereditary Hemorrhagic Telangiectasia: Risk Factors and Endoscopic Findings. Journal of clinical medicine. 2019; 9. 10.3390/jcm9010082. [DOI] [PMC free article] [PubMed] [Google Scholar]