Burden and Treatment of Achondroplasia: A Systematic Literature Review

Eighteen studies reported HRQoL outcomes for individuals with achondroplasia, of which 13 were conducted in a European setting (Fig. 2). The majority of studies included children only (n = 8) or a mixed population of children and adults (n = 7), with three measuring HRQoL in adults only (Table 1). Nineteen different instruments were used to elicit HRQoL data. The most commonly used was the Quality of Life in Short Statured Youth (QoLISSY) questionnaire, used in five studies. This was followed by the Pediatric Quality of Life Inventory (PedsQL) and the 36-Item Short Form Health Survey (SF-36), each used in four studies (Fig. 3). HRQoL was self-reported in 10 studies, caregiver-reported in one study, and both in seven studies. Utilities were assessed in two studies, one using the EQ-5D-5L scale and one using the 15-dimensional (15D) validated generic self-assessment instruments of HRQoL to measure utility indexes.

Two of the measured tools were condition-specific (QoLISSY and the Achondroplasia Personal Life Experience Scale [APLES]); the others were generic scales. QoLISSY is a tool that is scored from 0–100 (higher score indicates better HRQoL). It contains 22 items covering physical, social and emotional HRQoL, 10 items covering additional aspects of coping, four items covering general attitude to body height and additional items in the parent version covering child’s future and impact on parents [34]. APLES is an instrument that was developed based on the International Classification of Functioning-Children and Youth Version. It contains 21 items covering self-perception, friends, recreation, school and physical domains [35].

Six studies reported HRQoL data relating to an intervention for achondroplasia. The interventions included vosoritide (n = 1) [36], limb lengthening (n = 2) [37, 38], surgical procedures (not limited to limb lengthening) (n = 1) [39] and self-help/education seminars (n = 2) [34, 40]. Only the self-help and patient education interventions resulted in demonstrable benefits to patients’ HRQoL compared with scores for non-participants [34, 40]. Both studies were based in Germany, recruited participants from patient organisations and measured HRQoL using QoLISSY (details aforementioned). Both studies investigated self-help seminars that were designed following focus group discussions and a questionnaire for patients and their parents. They included eight modules covering physical, emotional, social and coping domains. In Rohenkhol 2016, across 58 children aged 8–17 years, mean (± SD) QoLISSY scores for participants after the intervention significantly increased by 5.12 (± 1.75; p = 0.003). This differed significantly from scores of 13 non-participants, which decreased by 2.94 (± 3.36) (p = 0.040). Similar increases were reported across all QoL domains, but the largest was social (+ 7.26), followed by physical (+ 6.52) then emotional (+ 5.01). The same study also compared patient- and parent-reported scores, finding that children reported a significantly more positive change in QoL compared to their parents (+ 4.10 vs. − 1.92; p = 0.001). Parents rated the change in emotional QoL as the worst, at − 4.27 [40]. Very similar findings were reported in Witt 2017, which included patients with achondroplasia aged 18–28. The total QoLISSY scores (reported by 61 patients and 44 parents) increased by 5.33 (± 1.55) in those that participated, whereas there was a reduction of 2.88 (± 2.68) for those who did not participate (p = 0.009). Similar gains were seen across, social, emotional and attitudes domains (+ 6.51, + 5.99 and + 5.20, respectively) [34]. The study also compared patients’ and parents’ perspectives (unrelated to any intervention) and found that patients rated their HRQoL significantly higher across all subscales of QoLISSY than their parents. Prior to intervention, the authors found that clinical, sociodemographic and psychosocial variables explained 49% of the variance of the QoLISSY total score, with attitudes towards body height identified as the most relevant predictor for HRQoL. It should be noted that both studies were conducted by the same research group and are therefore likely to have included overlapping patients.

A further 12 studies reported HRQoL unrelated to treatment. Four studies compared QoL in participants with and without achondroplasia [41‐44], four studies compared patient- and caregiver/parent-reported values [41, 44‐46], three compared subgroups of patients with achondroplasia (different height groups [47], age groups [48] and sexes [49]) and three studies made no comparisons [18, 50, 51]. In the four studies that compared caregiver/parent- and patient-reported values, where the age of patients ranged from 5–17 years, HRQoL was consistently judged to be lower by caregivers than patients, across all HRQoL scales [41, 44‐46]. This difference was significant in the three studies that reported results of statistical testing [35, 41, 45, 46]. One of these studies used a condition-specific HRQoL tool, APLES, and parents rated their child’s HRQoL significantly lower than the self-reported value for total score and across all domains apart from “interaction with others” [35]. Where scores for patients with achondroplasia were compared to those without achondroplasia, HRQoL was consistently lower in achondroplasia in both children and adult populations [41‐44]. In an Australian study that compared WeeFIM-II parent-reported scores for a population of children with achondroplasia aged 3–12 years, HRQoL was reported to increase with increasing age of the child [48]. In Matsushita 2019, a Japanese study that compared SF-36 scores for two height groups of children and adults (aged 10–67 years), mean physical component summary (PCS) scores were higher for those in the 140–159 cm group compared with those in the 100–139 cm group (49.42 ± 12.77 vs. 38.08 ± 17.20; p value not reported). However, mean mental component summary (MCS) scores were similar in both groups (52.47 ± 11.86 vs. 53.65 ± 10.66), indicating that there may be a stronger association between height and physical aspects of QoL than mental aspects [47]. Finally, in a US-based study of 25 adults aged 19–66 with achondroplasia, the mean self-reported PCS SF-36 score was non-significantly lower in female than male patients (35.2 ± 13.4 vs. 39.5 ± 17.2; p = 0.238) while the MCS score was non-significantly higher in female than male patients (42.3 ± 15.5 vs. 33.7 ± 14.5; p = 0.062) [49].

Five studies measured HRQoL using more than one separate tool [34, 36, 41, 44, 51, 52]. However, none of the studies aimed to compare tools in terms of their suitability for assessing HRQoL in achondroplasia. Instead, the same trends were reported across different tools. For example, the multinational LIAISE study found associations between physical domain and height Z-score in QoLISSY, and mobility and Z-score in WeeFIM [44]. In one study that reported results from two generic tools (KIDSCREEN: comprised of 10 items that assess general and subjective health and wellbeing; DISABKIDS: comprised of 10 items that assess the impact of chronic health conditions and two items that measure the impact of treatment) and one condition-specific tool (QoLISSY: details aforementioned), the authors concluded that QoLISSY was a reliable and valid tool to measure HRQoL in achondroplasia, based on the Cronbach’s alpha statistic and correlations with KIDSCREEN dimensions. However, they did not specifically comment on the suitability of this tool compared with the others [34, 41, 52].

Limited utility data were identified. One study, conducted in Finland, included a small number of adult participants (n = 8) with achondroplasia for which 15D utility values were measured and reported separately to other conditions [42]. Mean score was marginally lower in adults with achondroplasia compared with the control population (0.911 vs. 0.929). A second article (a conference abstract) reported EQ-5D-5L utility index scores for 74 adults with achondroplasia in the multinational Lifetime Impact of Achondroplasia Study in Europe (LIAISE) study [44]. The mean utility index score was 0.7 for adults with achondroplasia compared with 0.9 for the reference population.

Five out of six studies reporting caregiver QoL included parents of children or adolescents with achondroplasia; the other included any caregivers (Table 2). One study used the 8-Item Short Form Health Survey (SF-8) [45] and one used the Achondroplasia Parent Experience Measure (APEM) [53]. A third reported descriptive data on the personal impact on carers of children with achondroplasia rather than measuring caregiver burden using a specific QoL instrument [54]. Three studies reported the ‘Effect on parents’ subscale in the parent-report version of the QoLISSY questionnaire [36, 40, 52].

In the three studies where it was specifically evaluated, the QoL of parents of children with achondroplasia was detrimentally impacted [45, 53, 54]. In Witt 2019, a German cross-sectional study in 73 parents, parents reported significantly worse mental health compared with a reference population (mean SF-8 score 46.51 vs. 53.25; p ≤ 0.01), while physical health was not affected (mean score 50.50 vs. 50.30; p = 0.85) [45]. In two studies, caregiver emotional wellbeing was reportedly negatively impacted by achondroplasia [53, 54]. The impact of interventions on parent HRQoL was only reported in the vosoritide pivotal phase 3 trial 111-301, with a decrease of 2.50 on the QoLISSY parents subscale. However the confidence intervals were wide ranging [36].

Three studies published as congress abstracts reported costs or HCRU data in achondroplasia [54‐56] (Table 3). One study was US-based and reported hospital-related costs and length of stay [55]; the other two studies were international and reported HCRU.

The US-based study used 2017 data from the National Inpatient Sample and estimated total hospitalisation-associated costs for all patients with achondroplasia at approximately $40 million [55]. Notably, costs were equally contributed by adults and children ($19.7 million for adults, $19.9 million for children). Total mean per-patient inpatients costs were $19,959 (95% confidence interval [CI] $16,801–$23,118), an increase of $7789 for people with achondroplasia compared to the general population [55]. Average hospital length of stay was 6.8 days (5.7–8.0), an increase of 2.2 days compared to the general population. These data were published via a conference abstract and further details, such as main drivers of total costs, were not reported.

The multinational LIAISE study reported HCRU data based on a retrospective review of medical records from 186 patients aged 5–84 years over a minimum of five years [56, 57]. Data were stratified by age group and included inpatient admissions per patient per year (mean 2.5 [range 1.5–3.1]), medications reported per patient per year (mean 7.2 [range 4.4–14.7]) and number of different specialist visits per year (mean 3.7 [range 1.7–6.0]) [57]. Some HCRU categories appeared to be associated with age. For example, mean duration of stay per inpatient visit ranged from 3.7–6.7 days in the 0–5 to 21–30 age groups and from 11.7–21.0 in the 31–40 to 51–60 age groups. Meanwhile, the frequency of annual specialist visits was higher for age groups 0–5 and 11–15 (25.7 and 29.1, respectively, vs. a range of 3.0–11.3 for other age groups). However, no statistical testing for significance was conducted. An update from this study reported on surgical procedures and healthcare practitioner visits, as well as inpatient and outpatient stays, in the same period. Of 186 patients, 72.0% had undergone ≥ 1 surgical procedures [56].

A multinational cross-sectional survey of 660 parents/caregivers (Baratela 2021) found that, excluding Japan where GH is standard care treatment, two thirds of children with achondroplasia had a primary care visit every six months, which would often involve travel of > 60 miles to attend. Similar to findings from LIAISE, the frequency of visits was reported to decrease with increasing age (> 1 visit per year for > 90% of 0–2 year olds vs. 41–71% of 12–18 year olds) [54].

Forty unique studies (three RCTs, two extension studies, 17 non-randomised trials and 18 observational studies) reported on efficacy and/or safety of potential therapy options for achondroplasia. The majority of studies reported on either GH (n = 14) or limb lengthening (n = 21), while two trials, each with an extension study, investigated vosoritide, and one exploratory phase 1a trial investigated meclizine (Table 4). Most studies were conducted in the Asia-Pacific region (n = 18), followed by Europe (n = 13) (Fig. 2). Sample size was generally small, with 95% of the included studies including <100 patients and approximately 25% with a patient population <10. The most frequently reported clinical outcomes across all studies were change in standing height (n = 20) and growth velocity (n = 14).

A favourable effect on change in height was reported for all identified interventions (Fig. 4). Four clinical studies investigated vosoritide in children with achondroplasia aged ≥5 years [36, 58‐60]. In Study 111-301, a placebo-controlled RCT, patients receiving 15.0 μg/kg/day vosoritide for one year achieved a least-squares (LS) mean change in height Z-score of 0.27 (95% CI 0.18–0.36; p < 0.0001 vs. placebo) [36]. The mean change in standing height for the treated vs. untreated patients was 5.59 ± 1.06 vs. 3.93 ± 1.08 cm. The extension phase of the study (Study 111-302) demonstrated that benefits were sustained after two years of vosoritide treatment [60], with differences in LS mean change from baseline height of 3.34 cm (95% CI 2.76–3.93) and + 0.44 (95% CI 0.25–0.63) in height Z-score for vosoritide-treated vs. untreated patients. In Study 111-202, an open-label phase 2 study, and its extension, Study 111-205, children achieved an increase in mean standing height Z-score by 30, 42 or 60 months of treatment compared to baseline, with mean increase in height Z-score of 0.78 (± 0.70) after 60 months of treatment [58, 59].

Eight studies measured standing height in response to GH therapy over a time period of up to 10.7 years, with a significant favourable impact on standing height Z-score reported by five studies [36, 61‐64]. This ranged in an increase from baseline in standing height Z-score of +0.2 after one year of treatment [62] to +1.6 after five years of treatment [65]. The other three studies also reported positive changes in height Z-score following GH but did not report whether they were statistically significant [65‐67]. One study reported overall change in height following GH. Mean change in standing height was +2.8 cm after 9.3 (±2.5) years in females (p < 0.06 vs. baseline) and +3.5 cm after 10.7 (±4) years in males (p < 0.05 vs. baseline) [68].

Reported increase in standing height following limb-lengthening surgery varied considerably across 9 studies, from a mean of 5.7 [69, 70] to 20.5 cm [70], likely because of the use of different surgical procedures and population characteristics, such as age, in different studies. The studies included sample sizes of 3–29 patients at starting ages of 5–16.7 years (mean age at start of treatment was not reported in two studies) [69‐77]. The mean age at the start of treatment in the study reporting the greatest mean increase in standing height (20.5 cm) was 7.8 years [70]. Patients in five of these studies had undergone tibial and femoral limb-lengthening surgery [69‐71, 73, 76], and in three studies patients had undergone tibial lengthening alone [74, 75] [77]. One study did not clearly report procedures [72]. One study reported mean change in standing height at three separate time points after consecutive surgical procedures, demonstrating a cumulative effect with total increases of 5.7 cm after first tibial lengthening (n = 14), 6.5 cm after subsequent femoral lengthening (n = 8) and 8.7 cm after second tibial lengthening (n = 6) [69]. Ten limb-lengthening studies reported outcomes that were related to growth but not change in height, including change in tibial and femoral length separately, extent of elongation and change in arm span (data not shown) [38, 78‐87].

Out of 12 GH studies that reported AGV, four reported statistically significant increases following GH compared to baseline [61‐63, 88]. The greatest significant increase after one year of GH treatment was reported by Kanazawa 2003 (from 3.9 cm/year at baseline to 7.2 cm/year) [62, 89]. The smallest increase was from 3.9 cm/year at baseline to 4.6 cm/year after two years of GH in Tanaka 1998 [88]. A further six studies compared AGV following GH to AGV at baseline, but without testing for statistical significance [64‐66, 89‐91] (Fig. 5). In three GH studies that measured AGV at different time points, a trend towards tachyphylaxis was observed [64, 65, 88]. An RCT assessed two doses (low-dose and high-dose) of GH after 1 and 5 years of treatment. For both dose groups, mean change in growth velocity from baseline was lower after 5 years than after one year (low-dose group: 1.9 ± 1.2 cm/year at 1 year, − 0.08 ± 0.7 at 5 years; high-dose group: 3.6 ± 2.0 at 1 year, 0.8 ± 1.7 at 5 years) [65]. Similar findings were reported in two single-arm trials, with lower mean growth velocity after three years of treatment compared to one year [64, 88]. Again, statistical significance between the different time points was not assessed.

In vosoritide studies, a positive increase in AGV was observed. In the phase 2 studies, daily vosoritide treatment at a dose of 15 µg/kg resulted in sustained increases in AGV for up to 60 months (Fig. 5) [58, 59]. The benefit of vosoritide was further demonstrated by the results of the phase 3 randomized placebo-controlled trial (111-301), which showed a statistically significant improvement in AGV of 1.57 cm/year after 52 weeks compared to placebo [36]; furthermore, this AGV improvement was sustained after two years in the extension Study 111-302 (data not shown) [60].

Only two studies reported growth velocity in relation to limb lengthening. The first was a single arm trial that reported distraction rates ranging from 0.5–1.5 mm/day during tibial limb lengthening [77]. The second was a retrospective case-control study that presented change in growth velocity for patients following bilateral tibial lengthening. A statistically significant difference in growth velocity was not detected after one year (p = 0.53) but a decrease in mean growth rate of 59.5% was detected after two years (p = 0.03) for patients undergoing surgery (n = 23) compared with those under observation only (n = 12) [83].

In vosoritide studies, bone age was reported to have progressed normally [36, 58], indicating that vosoritide does not lead to premature bone ageing among children with achondroplasia. Findings were inconsistent in GH studies. After 2 years of GH therapy in one study, bone-to-chronological-age ratio was reported to decrease moderately, from 0.93 ± 0.13 to 0.90 ± 0.10, a positive result although statistical significance was not reported [66]. Another study found that GH therapy decreased mean bone mineral density (BMD) Z-score, though the effect appeared to lessen over time (baseline: 1.1; year 1: −0.6 ± 1.1; year 2: −0.21 ± 1.6; year 3: 0.04 ± 1.02) [64].

Nine out of 14 GH studies reported AEs. Six studies reported that no AEs occurred. However, it was not clear whether only serious AEs were considered and therefore mild events were not reported. In the remaining three studies, sleep apnoea, kidney failure and advancement of bone age (n = 2) were the only AEs observed [65, 66, 91]. In a phase 1a safety study on meclizine, no serious AEs were reported. Four out of six children experienced a low-grade AE in the group receiving one 25 mg tablet per day in the fasted state, while one out of six children experienced a low-grade AE in the group receiving two 25 mg tablets per day in the fed state [92]. As this study was only conducted over a 7-day period, longer follow-up would be needed to evaluate the reliability of this finding. Over studies 111-301 and 111-302 and their extensions, 98–100% of patients receiving vosoritide experienced an AE, most commonly injection site reaction and injection site erythema, at 73–86% and 68–86%, respectively [36, 58‐60]. Most AEs were mild, with a low proportion of serious AEs in both studies (5% in the treatment arm of the RCT; 7% in the placebo arm; 11% in the dosing/extension study). AEs were reported in 15 of 21 limb-lengthening studies, of which all were related to the surgical procedure. Most commonly reported were fractures and pin site/tract infections, and AEs related to soft tissue/nerve damage were also common [38, 70‐74, 76‐79, 84‐87, 93, 94].

No economic evaluations were identified for treatments for achondroplasia, highlighting the unmet need for studies exploring the cost-effectiveness of therapy options. In economic evaluations investigating therapies for other short stature conditions, drivers of cost-effectiveness results included dose of GH [95, 96] and utility values associated with height Z-scores and post-treatment quality-adjusted life years (QALYs) [97].

Across the clinical evidence base, only three of 40 studies were randomised, and of these, only one was placebo controlled [36, 61, 65]. The randomised studies were generally of high quality, with all three using intention-to-treat analysis, reporting similar baseline characteristics between arms and none reporting any unexpected drop-outs (Fig. 6). However, method of randomisation was not described in two RCTs [61, 65], and none provided details allocation concealment. Of the 17 interventional non-RCTs, the majority clearly described the measured outcomes, stated the objectives and provided estimates of the random variability in outcome data. However, in all 17 studies, the representativeness of patients to the entire population of children with achondroplasia from which they were recruited was unclear, as the studies did not report the proportion of the source population from which patients were derived. Of the 18 observational studies, 11 stated the objectives clearly, 13 described the main outcomes to be measured, and 14 clearly described the intervention of interest. However, the characteristics of patients was not described clearly by any study, highlighting a particular weakness in this area.

Thirteen guidelines on the management of achondroplasia and/or other short stature conditions were identified from targeted searches (Table 5). Nine of the 13 included guidelines were from the perspective of a single country, of which two had European perspectives (Wales [98], France [99]). Four had a US perspective [100‐103]; one Japanese [104], one South African [105], and one Australian [106]. Four guidelines had an international perspective [107‐110]. Common themes reported across multiple guidelines included monitoring and clinical evaluation, surgery and GH treatment. Monitoring was often recommended for specific complications, such as respiratory function monitoring in patients with thoracic spinal deformity. In achondroplasia, the Skeletal Dysplasia Management Consortium recommended a comprehensive history and physical examination be performed every two months to screen for foramen magnum stenosis (FMS) [102]. Furthermore, the Guidelines Development Committee for Achondroplasia from the Japanese Society for Pediatric Endocrinology strongly recommended foramen magnum decompression for managing spinal cord compression due to FMS [104]. GH recommendations for conditions such as growth hormone deficiency (GHD) included monitoring serum levels, measuring growth velocity and considering dose increases and reductions in various subgroups [101, 107, 108].

There was little consensus between the older guidelines, largely as they did not examine the same short stature conditions, use the same data collection methods or focus on the same aspects of treatment and management. However, a set of international management guidelines on achondroplasia by Savarirayan and colleagues were published in November 2021 [109]. These guidelines were developed by a group of 55 international experts using a modified Delphi process and provide consensus statements on many aspects of achondroplasia management and treatment across patients’ lifespan.

Key consensus statements from these guidelines are summarised in Fig. 7.

These statements align with the results from the clinical SLR, where it was found that there is limited evidence for the long-term efficacy of GH and its effect on body proportion ratios, which may be a more meaningful outcome for some patients. Given the recent approval of vosoritide, the therapy is expected to feature in more detail in future guidelines [26, 27].