FormalPara Key Summary Points

Why carry out this study?

For patients with atopic dermatitis (AD), pruritus is an extremely bothersome and debilitating symptom that affects quality of life (QOL) due to its effects on sleep, daytime functioning, and work productivity.

Nemolizumab administration to patients with AD can improve pruritus from as early as the second day after treatment initiation; skin findings are also improved.

To date, there has not been an in-depth evaluation of the impact of nemolizumab on QOL characteristics.

What was learned from this study?

Subcutaneous administration of nemolizumab ameliorated itchiness and skin signs, and thereby produced improvement in patient QOL across multiple patient-reported outcome measures, including sleep, interpersonal relationships, and ability to conduct social/work activities.

Administration of nemolizumab has wide-ranging therapeutic significance for patients with AD and pruritus, with benefits across multiple facets of this burdensome condition.

Introduction

Among chronic inflammatory dermatologic conditions, atopic dermatitis (AD) is the most common, posing a debilitating disease burden to both patients and society [1]. In a recent survey of adults diagnosed with AD, the most frequently reported signs and symptoms (pruritus, dryness, and erythema) were also the most bothersome for patients [2]. As a result, patients with AD suffer from reduced quality of life (QOL), even in comparison with patients with other chronic skin conditions [3].

Pruritus is known to have a particularly negative impact on patient well-being [4]. It is estimated that up to 80% of children and 90% of adults with AD suffer from sleep disturbances [5]. For the patients themselves, nocturnal itching causes a scratching response, reducing sleep efficiency and depth of sleep, and increasing wake time [6]. For parents of children with AD, levels of sleep disruption and cosleeping have been shown to directly impact family happiness [7]. For all those affected, either directly or indirectly, the resultant sleep deficit can lead to daytime somnolence and fatigue, and other physical and mental disturbances that have a deeply negative impact on QOL [8, 9]. In turn, daytime productivity, in terms of both work and normal activities, is reduced [10,11,12], adding to the large and multifaceted financial burden associated with AD [13, 14].

Nemolizumab is a humanized monoclonal antibody that targets interleukin (IL)-31 receptor A [15, 16]. The T-cell-derived cytokine IL-31 has been implicated in the induction of pruritus and skin inflammation, and in skin defects and barrier dysfunction [17]. Studies of nemolizumab administration to patients with AD have shown that pruritus is improved from as early as the second day after treatment initiation, and skin findings are also improved [18,19,20]. In a Japanese phase III clinical study, 60 mg nemolizumab administered subcutaneously every 4 weeks (Q4W), in combination with topical corticosteroids or calcineurin inhibitors (TCS/TCI), resulted in a greater reduction in pruritus associated with AD compared with placebo [visual analog scale (VAS) score −42.8% with nemolizumab versus −21.4% with placebo, difference −21.5%; p < 0.001] and in an improvement in cutaneous signs [Eczema Area and Severity Index (EASI) −45.9% with nemolizumab versus −33.2% with placebo, difference −12.6%] [18]. Moreover, efficacy is maintained without new safety concerns, even with long-term administration [21, 22], which is a key attribute in the treatment of this chronic condition.

However, although previous reports have shown that administration of nemolizumab is able to improve overall QOL, sleep [as measured by the Insomnia Severity Index (ISI)], and daily functioning [as measured by the Dermatology Life Quality Index (DLQI)], the data are limited [18, 23]. A detailed examination of the impact of nemolizumab on QOL characteristics has been lacking, and the relationships between QOL [as evaluated by patient-reported outcome (PRO) measures] and AD signs and symptoms have not been fully elucidated. Therefore, we used data derived from the recent Japanese phase III clinical study to report, in detail, the effects of nemolizumab 60 mg Q4W, administered in combination with TCS/TCI, on multiple PRO indicators and their relationship with improvements in pruritus and rash.

Methods

Study Design

Full details of this Japanese phase III, parallel-group, multicenter, clinical trial (registered as JapicCTI-173740) have been reported. In brief, it consisted of a 16-week double-blind randomized period (Part A) [18], after which patients could enter a long-term follow-up period (Part B), for a total treatment duration of up to 68 weeks [22]. The trial was conducted in accordance with the ethical principles of the Declaration of Helsinki, good clinical practice guidelines, and all other applicable regulatory requirements. The institutional review board at each participating center approved the protocol and related documentation. All patients, or their legal guardians, provided written informed consent prior to initiating study procedures and treatment.

Patients eligible for study participation were males or females aged ≥ 13 years with a body weight of 30–120 kg, and had an investigator-confirmed diagnosis of AD [24], with an inadequate pruritic response to TCS/TCI and oral antihistamines. This was defined using the following scales, with higher scores indicating greater severity: a score of ≥ 3 on the five-level itch scale [25], a pruritus VAS score of ≥ 50 [26], and an EASI score of ≥ 10 [27]. Full inclusion/exclusion criteria are provided in Supplementary Text S1.

Patients were randomly assigned (2:1 ratio) to receive either subcutaneous nemolizumab 60 mg Q4W or placebo in Part A [18]. All patients in Part B received nemolizumab 60 mg Q4W up to week 64, resulting in nemolizumab/nemolizumab and placebo/nemolizumab assessment groups [22]. Concomitant stable treatments for AD (medium-potency TCS/TCI and oral antihistamines) were to remain unchanged during Part A, and TCS of any potency could be used during Part B [18]. At the end of treatment, there was an 8-week follow-up period.

Endpoints

Details of the primary efficacy endpoint (the percent change in the weekly mean pruritus VAS score from baseline to week 16) have been published [18]. In the current post hoc analysis, we report details of the PROs completed by patients at study visits [ISI [28], DLQI [29], Patient-Oriented Eczema Measure (POEM) [30], 5-dimension 5-level EuroQoL (EQ-5D-5L) questionnaire [31], and the Work Productivity and Activity Impairment: Atopic Dermatitis questionnaire (WPAI-AD) [32]], and the correlations between these evaluation items and measures of AD signs and symptoms (pruritus VAS and EASI).

Data Analysis

The modified intention-to-treat population included all randomly assigned patients who received at least one dose of nemolizumab or placebo and had data available for evaluation; missing data were not imputed.

Categorical variables were described using number and proportion of patients, and continuous variables using mean, standard deviation, and standard error of the mean. For this post hoc analysis, between-group differences in the QOL measures were calculated using the t-test, with p < 0.05 indicating nominal significance. The absence of symptoms on each of the ISI, DLQI, and POEM scales was calculated using the number of patients with a score of 0 for each item measured. Between-group differences in the absence of symptoms were calculated using Fisher’s exact test. Correlations between the QOL measures and VAS or EASI were depicted using scatterplots. The Pearson correlation was used to evaluate the linear relationship between the two continuous variables, and the Spearman correlation coefficient was calculated to assess the ranked strength of the relationship. All statistical analyses were performed with the use of SAS software, version 9.4 (SAS Institute, Inc., Cary, NC, USA).

Results

Study Population

A total of 215 Japanese patients were enrolled into the study and were randomly assigned to receive either nemolizumab (n = 143) or placebo (n = 72) in Part A; all patients received at least one dose of their allocated treatment and were included in the modified intention-to-treat population [18].

Baseline demographics and disease characteristics are summarized in Table 1. There were no notable differences in the key attributes between the nemolizumab and placebo groups, and AD severity at baseline was comparable between the groups.

Table 1 Patient demographics and clinical characteristics at baseline (modified intention-to-treat population)
Full size table

PRO Outcomes

Pruritus VAS and EASI total scores at baseline, week 16, and week 68, and percentage change from baseline at weeks 16 and 68 are presented in Table 2. The change from baseline to week 16 in the ISI, DLQI, POEM, and WPAI-AD scores in patients in the nemolizumab treatment group, compared with the changes in the placebo-treated group, are also reported in Table 2. The proportions of patients at week 16 who reported a score of 0 for each item on the ISI, DLQI, and POEM, and who reported no pain/discomfort on the EQ-5D-5L, are shown in Fig. 1.

Table 2 Summary statistics for the pruritus VAS, EASI, ISI, DLQI, POEM, and WPAI-AD assessments at baseline, week 16, and week 68 (modified intention-to-treat population)
Full size table
Fig. 1
figure 1

Proportions of patients reporting 0 for each symptom at week 16 (modified intention-to-treat populationa). DLQI Dermatology Life Quality Index, EQ-5D-5L 5-dimension 5-level EuroQoL questionnaire, ISI Insomnia Severity Index, POEM Patient-Oriented Eczema Measure. p-Values for nemolizumab versus placebo were calculated using Fisher’s exact test. *p < 0.05, **p < 0.01. aAnalysis was conducted only for patients with a score of 1 or more for the selected symptom at baseline

Full size image

Sleep

There was a significant difference (p < 0.01) in the change from baseline to week 16 in the ISI scores in patients in the nemolizumab treatment group compared with the changes in the placebo-treated group (Table 2).

An ISI score of ≥ 8 has previously been reported to indicate sleep dysfunction [28]; in our analysis around 80% of patients had some level of sleep dysfunction (score ≥ 8) at baseline (Fig. 2A). By week 16, almost two-thirds of nemolizumab-treated patients (65%) had no clinically significant insomnia (score ≤ 7) compared with 33.8% of placebo-treated patients, indicating that nemolizumab improved sleep quality to a greater extent than placebo. Improvements were maintained through to week 68. Patients in the placebo group who received nemolizumab after week 16 also showed improvement in the ISI score between week 16 and week 68. Improvements in ISI were observed as early as 2 weeks after initiating nemolizumab treatment (i.e., at the time of the first post-treatment ISI evaluation, Supplementary Table 1).

Fig. 2
figure 2

Change in scores according to treatment group for A ISI, B DLQI, and C POEM. DLQI Dermatology Life Quality Index, ISI Insomnia Severity Index, POEM Patient-Oriented Eczema Measure

Full size image

Few patients were satisfied with their sleep at baseline, with ~ 80% reporting difficulties with falling asleep, interference with daily functioning, and worries about their sleep. By week 16, significantly more patients in the nemolizumab group versus the placebo group had a score of 0 for difficulty falling asleep (41.6% versus 13.1%, nominal p < 0.01), difficulty staying asleep (45.4% versus 10.9%, nominal p < 0.01), interference with daily functioning (31.8% versus 9.1%, nominal p < 0.01), and worries about their sleep (36.3% versus 11.7%, nominal p < 0.01), indicating that nemolizumab was able to improve multiple aspects of sleep and daily activity (Fig. 1). The difference between the nemolizumab and placebo groups in the proportions of patients who had difficulty staying asleep was particularly noticeable. By week 68, nearly half of patients reported that they had no difficulties falling asleep, staying asleep, or waking up too early, or in the extent to which their sleep issues were noticeable to others.

Quality of Life

A significant difference (p < 0.01) in the change from baseline to week 16 in the DLQI scores was observed (Table 2) between the two treatment groups.

The DLQI is interpreted using a threshold of ≥ 6 representing a moderate effect and ≥ 11 representing a very large effect [29]. At baseline, more than 90% of patients reported moderate or severe effects on QOL (DLQI score ≥ 6, Fig. 2B). By week 16 and week 68, this had decreased to ~ 50% and ~ 20%, respectively, for patients in the nemolizumab-treated group. Moreover, while no patients reported a DLQI score of ≤ 1 at baseline (no effect on QOL), almost one-third of nemolizumab-treated patients (30.5%) had a score of ≤ 1 by week 68. The effect of nemolizumab on DLQI was rapid, with improvements observed as early as week 1 (Supplementary Table 1).

Baseline DLQI scores indicated that ~ 80% of patients had itchy or sore skin, were embarrassed or self-conscious, had problems with daily activities (shopping, or working in the house/garden) and in their choice of clothing, and felt that their AD impacted relationships and work or study and that treatment was problematic (in terms of time taken and messiness of the procedures). By week 16, the proportion of patients with a score of 0 was significantly higher in the nemolizumab group versus the placebo group for interference with daily activities such as shopping, housework or gardening (45.2% versus 18.6%, nominal p < 0.01) (Fig. 1). By week 68, 60–70% of nemolizumab-treated patients reported no impediment to daily activities, relationships, or work/study (Supplementary Fig. S1).

Signs and Symptoms of Atopic Dermatitis

As shown in Table 2, there was a significant difference (p < 0.01) in the change from baseline to week 16 in the POEM scores between the two treatment groups.

POEM scores are categorized into severity strata, with a score of ≥ 17 indicating severe eczema and a score of ≥ 25 indicating very severe eczema [33]. A large proportion of patients (79.6%) assigned to nemolizumab treatment had severe or very severe eczema at baseline (score ≥ 17, Fig. 2C). This proportion decreased to 30.7% by week 16 and 15.9% by week 68. Although just 1.4% of patients in the nemolizumab group had a POEM score of ≤ 7 at baseline (mild or no eczema), this had risen to 50.7% by week 68. Like DLQI, POEM improvements were observed as early as week 1 (Supplementary Table S1), indicating rapid amelioration of subjective symptoms following initiation of nemolizumab treatment.

At baseline, > 80% of patients reported the presence of each symptom on at least 1 day per week. As shown in Fig. 1, by week 16, significantly more patients in the nemolizumab-treated group versus the placebo group had 0 days per week of nighttime sleep disturbance (50.8% versus 16.9%, nominal p < 0.01), bleeding skin (43.4% versus 7.5%, nominal p < 0.01) or weeping skin (29.4% versus 12.3%, nominal p < 0.05). The particularly large difference between the groups, in the proportions of patients with 0 days of bleeding skin by week 16, indicated the considerable effectiveness of nemolizumab on this aspect of AD. At week 68, the proportion of patients in the nemolizumab/nemolizumab group with 0 days per week of nighttime sleep disturbance or bleeding skin had increased to 70.5% and 50.7%, respectively (Supplementary Fig. S1).

Work Productivity

Although the loss of working hours due to AD was just 2–3% at baseline (WPAI-AD question 7), both labor productivity and activities of daily living were compromised by AD, with respective reductions of 46.7–47.5% (WPAI-AD question 8) and 54.3–54.9% (WPAI-AD question 10) at baseline (Table 2). Based on a full-time work schedule of 40 h per week, this is equivalent to a loss of ~ 20 h per week in labor productivity or daytime activities. After the initiation of nemolizumab treatment, rates of both labor productivity and activities of daily living continuously improved over time from baseline to week 68 in the nemolizumab/nemolizumab group and from week 16 to week 68 in the placebo/nemolizumab group. By week 68, labor productivity and daily activities were reduced by only ~ 20% (equivalent to a loss of ~ 8 h per week).

Correlations between Pruritus and PRO Outcomes

Correlations between the pruritus VAS and each PRO-related indicator are shown in Supplementary Fig. S2, and correlations between the EASI score and each PRO-related indicator are shown in Supplementary Fig. S3.

We found moderate correlations between the pruritus VAS and each PRO, suggesting that patients with reduced pruritus also had improved QOL. In particular, the pruritus VAS score was correlated with the POEM score (Pearson’s correlation coefficient ρ = 0.5729) and with the WPAI-AD question 10 (percent activity impairment, ρ = 0.5148) (Supplementary Fig. S3).

Moderate to weak correlations were also observed between the EASI score and each PRO (Supplementary Fig. S2), suggesting that reduction of eczema was associated with improvements in QOL for patients. An additional analysis including only those patients with more severe EASI scores (≥ 16) at baseline produced stronger correlations (Supplementary Fig. S4).

Safety

The safety data from this study have been published [18]. Briefly, the incidence of adverse events was similar in the nemolizumab and placebo groups up to 16 weeks (both 71%). Long-term use of nemolizumab 60 mg Q4W with concomitant TCS/TCI also resulted in a favorable safety profile with no new concerns [22].

Discussion

The results of this post hoc analysis from the Japanese phase III clinical trial of nemolizumab plus TCS/TCI demonstrated that subcutaneous administration of nemolizumab 60 mg Q4W produced improvement in the signs and symptoms of AD, and in patient QOL, as measured by multiple PROs and outcome assessments. The results also confirmed that nemolizumab treatment can improve daily functioning for patients with AD. Studies of other treatment agents have also reported improvements in QOL and functioning following initiation of treatment for AD [34,35,36], but our data have specifically shown that long-term administration of nemolizumab can improve interpersonal relationships and ability to conduct social activities (based on DLQI and WPAI-AD item scores), in addition to reducing pruritus and rash. This suggests that nemolizumab had wide-ranging therapeutic significance for affected patients, with benefits across multiple facets of this burdensome condition.

We have previously shown that nemolizumab-treated patients obtain sleep benefits (measured by ISI) as early as 2 weeks after initiating treatment [18], with improvements in sleep maintained for up to 68 weeks of treatment [22]. It is known that sleep disruption is a particular issue for patients with AD [5], with poor sleep listed by patients and their families in a recent survey as the third most problematic symptom they endure [37]. Notably, in the same survey, the most burdensome symptom was pruritus [36]. Due to the importance of this issue, there is a great deal of interest in how treatment for AD might also improve levels of sleep and fatigue, with the recent publication of several studies investigating changes in sleep disturbance following pharmacologic treatment [34, 38, 39]. The results of the current analysis confirmed the benefits of nemolizumab treatment on sleep parameters. Notably, pruritus improved from 2 days after nemolizumab administration, and had decreased by 30% after 15 days of treatment, while for each ISI item, significant improvements in difficulty falling asleep, staying asleep, and daily functioning were also observed from an early stage (from 2 weeks), following initiation of nemolizumab treatment. This suggests that reducing pruritus helps to maintain normal sleep patterns at night, and thereby improves daytime activity. Moreover, based on the POEM items for “number of nights of disturbed sleep” and “number of days of bleeding skin,” a high proportion of nemolizumab-treated patients had a score of 0 at both week 16 and week 68. We consider that this improvement is related to the amelioration of pruritus by nemolizumab; suppression of the itch-scratch cycle is likely to prevent symptom worsening, and contributes to the maintenance of a higher QOL.

There were no statistically significant improvements in either the DLQI item “extent of itchy, sore, painful, stinging skin,” or the POEM item “number of days of itchy skin.” The exact reasons for this remain to be confirmed, but it seems likely that although nemolizumab suppressed pruritus, it was difficult to eliminate itching completely. Thus, it is possible that the DLQI and POEM, which evaluate the frequency of pruritic events, were insufficient to show any significant improvement. However, although there were no reported improvements in “sore, painful, stinging skin” on the DLQI, we did observe an improvement in the number of patients who reported no pain/discomfort at week 16 on the EQ-5D-5L in the nemolizumab treatment group, compared with the placebo group. Although previous publications have suggested that administration of IL-31 to mice does not induce pain behavior [40], and that IL-31 receptor A-deficient mice show no differences to heat response compared with wild-type mice [41], it would appear that IL-31 receptor A deficiency affects the analgesic effect of morphine [41], suggesting that there is a relationship between IL-31 and pain. Our data would support this inference, although further studies would be needed to provide a definitive conclusion on this point.

AD also contributes to work impairment, with clear associations reported between pruritus and rash severity and reduction in productivity [11, 42]. Currently, there are no defined and validated minimally clinically important differences for WPAI-AD outcomes, although those for the work productivity loss and the activity impairment domains in patients with psoriasis have been estimated to be 20% [43]. Our data indicated that the disability rate at baseline in patients with AD was around 50%, and that nemolizumab treatment was able to reduce this disability rate by half for both labor productivity and activities of daily living, thereby increasing the ability of patients with AD to resume a more normal life and potentially reducing the financial burden of the disease. A prior publication showed that an increase in the severity of AD affected WPAI scores [11]; therefore, the finding that nemolizumab treatment suppresses itching and increases labor productivity by 50% is a meaningful outcome for patients.

The correlations between levels of pruritus and rash and the PROs evaluated in our study are of particular interest, since the scatter plots confirm that improvements in itching and skin inflammation observed after administration of nemolizumab plus TCS/TCI were associated with improvements in sleep quality and daytime activity. It has previously been reported that pruritus affects QOL in Japanese patients with AD [42]. Another recent post hoc analysis, using pooled data from two phase III studies of crisaborole, indicated that the contribution of itch amelioration is greater than that of improvement in rash in determining QOL [44]; thus, the impact of nemolizumab in reducing pruritus would be expected to lead to improvements in various aspects of QOL. Notably, correlations between rash and PROs were weaker than those between pruritus and PROs in our analyses. This may be due to the fact that patients with a relatively low baseline EASI score of ≥ 10 were eligible for inclusion in our study. When the correlation analysis was restricted to the ~ 80% of the patients who had an EASI score of ≥ 16 [35] at baseline, correlations between EASI and PROs at week 16 were observed to be stronger, compared with the overall population. Thus, we consider that the minority of patients with a less severe EASI score at baseline will have impacted the correlations for this measure.

Limitations

The main limitation associated with this analysis is the post hoc design, which means that statements of significance are nominal and may require further corroboration in adequately powered, prospective studies. Although slightly more adult females than males appear to be affected by AD, and it has been reported that females may be more susceptible to pruritogens [45], the current analysis did not address PROs according to sex. However, treatment with nemolizumab has been shown to be effective regardless of sex (Maruho Co., Ltd., data on file); thus, no sex-based differences in PRO outcomes were anticipated. An additional limitation is that only those patients with electronic diary compliance rates of > 75% during the screening period were included in the clinical study. Those with compliance rates during screening of < 75% were not assigned to treatment. As such, there is no information on clinical or PRO outcomes for these patients. Now that nemolizumab is approved for use in Japan, it is anticipated that future real-world assessments may provide additional information to fill in data gaps for patient populations who were excluded from the clinical trial. It is also well known that TCS have a strong anti-inflammatory effect, which could have affected PRO assessments and confounded our data. However, data from our previously published long-term analysis of this study [22] found that the amount of TCS used decreased as the study continued, suggesting that any impact on PROs would also diminish over time, thereby reducing any confounding effects. In addition, we have previously reported that there was no notable difference in the slope of the rate of change of the EASI score around week 16, even though TCS usage could change at this timepoint, indicating that changes in TCS usage did not have any clinically meaningful impact on EASI improvement [22]. Moreover, in a second long-term study of nemolizumab, despite the fact that any potency TCS were permitted to be used from the start of the study, the rate of change in the EASI score was comparable to that in the current study (in which TCS usage was limited in Part A and unrestricted in Part B), indicating little to no impact of TCS on inflammation [22]. Finally, it should be noted that this study targeted patients with AD who were unable to obtain sufficient pruritus symptom relief from TCS/TSI treatment; as such, we can hypothesize that these patients have already obtained any possible QOL benefit associated with the anti-inflammatory effects of TCS, and we can assume that continuation of a stable dose of TCS is unlikely to have any new impact on PRO outcomes. Overall then, we can find no evidence for major confounding effects resulting from the use of TCS, and consider that the beneficial effects of treatment reported herein result entirely from the use of nemolizumab.

Conclusion

Using data from a Japanese phase III clinical trial of patients with moderate-to-severe AD, we found that subcutaneous administration of nemolizumab 60 mg Q4W was able to improve pruritus and rash, and improve QOL across multiple PROs. Importantly for patients, the amelioration of pruritus and skin signs by nemolizumab also resulted in improvements across a spectrum of areas measurable by PROs, including sleep, interpersonal relationships, and the ability to conduct normal daily activities.