ABSTRACT

Objective:

Both body weight (BW)- and body surface area (BSA)-based dosing regimens have been recommended for growth hormone (rhGH) replacement. The aim was to compare the two regimens to determine if either resulted in inadequate treatment depending on anthropometric factors.

Conclusion:

Dosing based on BW may be preferable in both prepubertal and pubertal children who do not show adequate growth responses. In prepubertal children, relatively lower doses calculated based on BW rather than BSA provide similar efficacy at lower costs.

Results:

The study included 60 patients (18 girls, 30%). Thirty-six (60%) patients were classified as HAG. The ratio of dosing based on BW-to-BSA was positively correlated both with the ages and body mass index (BMI) levels of the patients, leveling off at the age of 11 at a BMI of 18 kg/m2. The relative dose estimations (over- and under-dosed groups) differed significantly between the patients classified as HAG or NHAG. Fifty-six percent of NHAG compared to 44% of HAG patients received relatively higher doses, while 79% of HAG compared to 21% of NHAG received relatively lower doses (p=0.006). When the patients were subdivided according to their pubertal status, higher doses were administrated mostly to the pubertal patients in both the NHAG and HAG groups. In the pre-pubertal age group, 73% of NHAG compared to 27% of HAG received relatively higher doses, while 25% of NHAG compared to 75% of HAG received relatively lower doses (p=0.01).

Methods:

The retrospective study included children diagnosed with idiopathic isolated growth hormone deficiency. BW-based dosing in mcg/kg/day was converted to BSA in mg/m2/day to determine the equivalent amounts of the given rhGH. Those with a BW-to-BSA ratio of more than 1 were allocated to the “relatively over-dosed group”, while the remaining patients with a ratio of less than 1 were assigned to the “relatively under-dosed” group. Patients with a height gain greater than 0.5 standard deviation score (SDS) at the end of one year were classified as the height gain at goal (HAG), whereas those with a height gain of less than 0.5 SDS were assigned as the height gain not at goal (NHAG).

What is already known on this topic?

Drug doses calculated based on body weight and body surface area may be different under certain circumstances.

What this study adds?

In children younger than 11 years of age with body mass index levels less than 18 kg/m2, growth hormone dosing based on body weight may be preferable.

Introduction

Dysfunction of the growth hormone (GH)-insulin-like growth factor-1 (IGF-1) axis may result in varying degrees of growth failure and a variety of other pathological clinical features, including central obesity, loss of lean muscle mass, osteoporosis, deterioration of metabolic profile, and decreased cardiac function (1,2,3). The diagnosis of growth hormone deficiency (GHD) is accomplished by combining medical history, auxological measures, biochemical markers, and radiological imaging (2,4).

The standard treatment for GHD is recombinant human growth hormone (rhGH) replacement, which is customized to each child’s body weight (BW) or body surface area (BSA) (5). Both BSA- and BW-based dose regimens have been recommended for rhGH replacement, on the assumption that they are equally effective (5,6). Yet, some countries still choose one regimen over the other (7,8,9,10,11).

Nevertheless, under certain circumstances, the differences between the two dosing regimens may become apparent (4,7,8,12). For instance, it has been demonstrated that BW-based dosing considerably underestimates the necessary treatment in individuals with weights of less than 30 kg (6). In contrast, due to disparities in drug clearance, obese patients are at risk of over-exposure when BW-based dosing is used, and of underexposure when BSA-based dosing is preferred (13). Hence, alternate rhGH dosing may be utilized for certain patient groups to boost effectiveness and/or to decrease toxicity (13,14).

In general, BSA-based regimens are favored for antineoplastic medications, whereas BW-based regimens are favored for cardiovascular, central nervous system, and anti-infective treatments (13). BSA-based dosing has been considered to be more closely associated with total body water, extracellular body fluid, total clearance, liver volume, and renal function (15,16). Given that rhGH is predominantly metabolized in the liver and kidneys, and the kidneys account for around 60-90% of the clearance, BSA-based dosing seems to be a more favorable method for rhGH replacement (7,17,18). However, there is insufficient data to support one over the other (5).

There is limited data (7,8) comparing the effectiveness of different rhGH dosages with regard to BSA versus BW. Homogenous studies on BW and BSA-based dosing strategies are needed. This retrospective study was designed to compare the rhGH doses in BW versus BSA in children diagnosed with idiopathic isolated GH deficiency (IGHD) who were not obese. Moreover, based on the growth responses of the patients over the first year of treatment, it would be evaluated whether either of the two regimens would result in higher or lower treatment under different conditions of age and patient anthropomorphic characteristics.

Methods

Patients

The retrospective study included children aged 1-18 years from two different centers. Individuals with obesity, defined as a body mass index (BMI) standard deviation (SD) scores ≥2, genetic anomalies, scoliosis, chronic diseases, such as diabetes mellitus or celiac disease, a history of significant trauma, low birth weight, neoplasia, brain tumor, or intracranial radiation were excluded from the study. Children with short stature who had IGHD and were treated with rhGH for at least one year between 2017 and 2022 were included. GHD was suspected in the presence of short stature (<-2 SDS) or growth deceleration (velocity <25% of corresponding chronological age), and diagnosed when serum peak GH concentration was less than 7 ng/mL in two different GH stimulation tests (clonidine, insulin tolerance test, and levodopa) (2,9). Isolated deficiency is defined as the presence of a solitary pituitary hormone deficiency. Each child received 25-35 mcg/kg/day of rhGH replacement (4). The changes in height velocity and height SD scores were evaluated to assess treatment efficacy while IGF-1 and insulin-like growth factor binding protein-3 (IGFBP-3) levels were monitored in order to avoid excessive dosing to ensure safety of the treatment (2,4,19,20). Clinical and laboratory data were monitored every 6-12 months to adjust rhGH doses (mg) (2,4). To exclude concomitant pathologies, each patient underwent pituitary MRI at the start of the therapy.

Data Collection

The following clinical parameters were recorded: age (years); gender; pubertal status [according to Tanner and Whitehouse (21)]; bone age [calculated according to the Greulich and Pyle (22) Atlas; height [measured with a sensitivity of 0.1 cm, using a Harpenden stadiometer, (cm)], weight [measured using a scale with a sensitivity of 0.1 kg, (kg)], BMI (kg/m²), target height (mother’s height + father’s height)/2±6.5, (cm)], predicted adult height [calculated according to the Roche et al. (23) method, (cm)], and the respective SD scores [calculated according to Turkish standards (24)]. The IGF-1/IGFBP-3 levels and the respective SD scores were recorded (24). The prescribed rhGH doses based on BW were also recorded. All data, which was re-evaluated every six months, was recorded.

Design of the Study

For the purpose of this paper, BW-based dosing in mcg/kg/day, which is routinely employed in our clinical practice, was converted to BSA in mg/m²/day. Assuming that the average BW of a child with a BSA of 1 m² is 28 kg (7,8), all doses were separately converted to equivalent BSA formats. Hence, the routinely prescribed doses of 25, 30, and 35 mcg/kg/day were found to be equivalent to 0.7, 0.8, and 1 mg/m²/day, respectively.

Then the BSA of each patient was calculated separately using the following empirical formulas:

$\mathrm{i}. \mathrm{Costeff}’\mathrm{s} \mathrm{Formula} \left(25\right): \mathrm{BSA} \left({\mathrm{m}}^2\right) = \frac{4 x weight\mathit{ }\mathit{\left(}kg\mathit{\right)}\mathit{+}7}{weight\mathit{ }\mathit{\left(}kg\mathit{\right)}\mathit{+}90}$

$\mathrm{ii}. \mathrm{Mosteller}’\mathrm{s} \mathrm{Formula} \left(26\right): \mathrm{BSA} \left({\mathrm{m}}^2\right) = \sqrt{\frac{weight\mathit{ }\mathit{\left(}kg\mathit{\right)}x\mathit{ }height\mathit{ }\mathit{\left(}cm\mathit{\right)}}{3600}}$

Finally, initially prescribed doses based on BW (mcg/kg/day) and the hypothetically calculated equivalent doses based on BSA (mg/m²/day) were calculated for each patient to be given as milligrams per day (mg/day).

Stratification of the Patients

Patients were divided into two groups based on their height increase over one year. The change in height SD score was determined by subtracting the height SD score measured at the beginning of treatment from the height SD score measured after the first year of rhGH treatment. Based on Bang criteria (27), those with a height gain greater than 0.5 SD score at the end of one year were classified as height gain at goal (HAG), whereas those with a height gain of less than 0.5 SD score were classified as height gain not at goal (NHAG).

Patients were also divided into two groups based on their actual (BW-based) and estimated (BSA-based) rhGH doses in mg. Those with a BW-to-BSA ratio of more than 1 were allocated to the “relatively over-dosed group” (n=32), while the remaining patients with a ratio of less than 1 were assigned to the “relatively under-dosed” group (n=28).

Ethical Approval

This study was approved by the Ethical Committee of Dokuz Eylül University Faculty of Medicine (approval number: 2022/42-14, date: 28.12.2022) and performed according to the principles of the Declaration of Helsinki. Informed consent to participate in the study was obtained from all participants (or their parents or legal guardian in the case of children under 16).

Statistical Analysis

Statistical analyses were performed using Statistical Package for the Social Sciences, version 24 for Windows (IBM Inc., Armonk, NY, USA). The homogeneity of the data obtained in the study was tested using Shapiro-Wilk and Kolmogorov-Smirnov tests. Numerical variables were not normally distributed and so non-parametric presentation was used; median [interquartile range (IQR)], unless otherwise stated. The correlation between the actual and estimated doses was assessed with Spearman’s correlation test (r_s). Categorical variables were analyzed by chi-square or Fisher’s exact test. All tests were two-tailed, and a p value of less than 0.05 was taken as statistically significant.

Results

The study included 60 patients with IGHD, of whom 18 (30%) were girls, and the whole cohort had a median (IQR) age of 11.9 (3.8) years. Table 1 summarizes the characteristics of all patients and the comparison of patients with HAG vs NHAG. Of the 60 patients, 36 (60%) were classified as HAG after one year of treatment. Overall, the median (IQR) dose administered per kg BW (mcg/kg/day) was reduced significantly over 1-year period [30 (4) and 27.6 (7) mcg/kg/day; p=0.007]. While the doses were similar in the prepubertal group [30 (5) and 28 (7), p=0.29, respectively], they were significantly reduced in the pubertal group [30 (1.5) and 26 (7), p=0.002, respectively]. The two groups classified as HAG and NHAG were not significantly different in terms of sex, ages at the start of treatment, puberty status, rhGH doses, target height, predicted adult height, and SD scores for weight, BMI, IGF-1, and IGFBP-3 (Table 1). The follow-up of the SD scores for weight and height, along with IGF-1 and IGFBP-3 levels are presented in Figure 1. The SD scores for IGF-1 levels were in the reference ranges at the end of the first year of treatment. IGF-1 levels were not correlated with the prescribed doses (r_s=0.164, p=0.22; r_s=0.14, p=0.3; r_s=0.14, p=0.3, according to BW and BSA calculated either by Costeff’s and Mosteller’s formulas, respectively).

The estimated daily doses calculated for BSA using Costeff’s and Mosteller’s formulas were strongly correlated [(r_s)=0.974, p<0.001]. The actual daily doses given, based on BW and the estimated doses calculated according to BSA were also strongly correlated (Spearman’s correlation (r_s)=0.990, p<0.001; (r_s)=0.977, p<0.001, BSA calculated with Costeff’s and Mosteller’s formulas, respectively). The median BW-to-BSA was 1 (0.2), with a full range of 0.65 to 1.34, while BSA-to-BW ratio ranged from 0.75 to 1.53. The ratio of the dose given based on BW and the dose calculated according to BSA were positively correlated both with the ages and BMI of the patients for both Costeff’s formula, (r)=0.814, p<0.001, (r)=0.776, p<0.001 and Mosteller’s formula (r)=0.747, p<0.001, (r)=0.797, p<0.001) (Figure 2). As shown in Figure 2a, b, the ratio of BW-to-BSA was equal to 1 at the age of approximately 11 years with a BMI of 18 kg/m². The slopes and the intercepts calculated for the best-fit lines for both Costeff’s and Mosteller’s formulas yielded similar results (age: Costeff’s formula: Y=0.03331*X + 0.6254; Mosteller’s formula: Y=0.03587*X + 0.6009; BMI: Costeff’s formula: Y=0.03250*X + 0.4222; Mosteller’s formula: Y=0.03939*X + 0.3043).

Table 2 shows the number and percentages of patients according to growth responses (HAG, NHAG) and relative dose estimations (relatively over- and under-dosed groups). The relative dose estimations (relative over- and under-dosed groups) differed significantly between the patients classified as HAG or NHAG. Fifty-six percent of patients in the NHAG group compared to 44% of patients in the HAG group received relatively higher doses, while 79% of patients classified as HAG compared to 21% of patients classified as NHAG received relatively lower doses (p=0.006). When the patients were subdivided according to their pubertal status, the results showed that higher doses were administered mostly to the pubertal patients in both NHAG and HAG groups (10/18; 56% and 11/14; 79%, respectively). In the pre-pubertal age group, 73% of patients classified as NHAG compared to 27% of patients in the HAG group received relatively higher doses, while 25% of patients classified as NHAG compared to 75% of patients classified as HAG received relatively lower doses (p=0.01). In the pubertal groups, patients in the NHAG and HAG groups received comparable doses (p=0.125) (Table 2).

Discussion

In this study, differences between the BW- and BSA-based dosing methods have emerged in children. We have shown that rhGH dosage calculations based on BW compared to BSA may result in the administration of relatively higher or lower doses, depending on the ages and BMIs of the patients, which may be particularly important in patients with good growth responses.

The notion that BW- and BSA-based dosages are equivalent has been examined by various studies in several disciplines, and the potential risks of over and undertreatment have been established (6,28,29,30,31). Both strategies were assumed to be equally effective in treatment, with significant differences only observed at the extremes of weight and in very young patients (32). In contrast to the general belief of equal efficacy, Hughes et al. (7) demonstrated that even slight increases in BW-based doses could correspond to higher values when converted to mg/m²/week. The results of the present study indicate that the difference between BW- and BSA-based dosing increased proportionally as patients’ age and BMI values increased. The actual and estimated doses were equal at the age of approximately 11 years with a BMI level of 18 kg/m². Thus, older patients with higher BMIs would be given higher doses if BW-based methods were chosen over BSA-based calculations.

Due to the variations in the pharmacokinetics of medications resulting from changes in growth and maturity, dosage recommendations for children are often subdivided into age categories of 2-6 years, 6-12 years, 12-18 years, and 18-21 years (33). Likewise, the Pediatric Pharmacy Advocacy Group (33,34) also recommends BW-based dosing for children weighing less than 40 kg. Even though these strategies have not been generalized for patients who are receiving rhGH treatment, different dosing strategies have also been explored among girls with Turner syndrome (8). Similar results were seen in the homogenous group of patients with IGHD with similar characteristics in the present study, also suggesting that different efficacy and safety profiles may result for different age groups.

Differences were also found between the estimations based on BW and BSA when patients were stratified by response to treatment in terms of height gained. Among patients in the HAG or NHAG groups, the percentages of patients who would be have been given relatively higher or lower doses of rhGH differed significantly. Fifty six percent of all patients classified as NHAG, most of whom were pubertal, received higher doses when using BW-based in comparison to BSA-based calculations. Furthermore, almost three-quarters of pre-pubertal patients (73%) classified as NHAG were given relatively higher doses using BW-based calculation. This could suggest that this method based on BW would be preferable in both prepubertal and pubertal groups with inadequate height gain, since BSA-based estimates would result in the administration of relatively lower doses to those with poor growth responses. However, these associations should be interpreted carefully. Although the differences were not significant, patients in the NHAG group had relatively shorter target heights and older bone ages, with statistically significant lower height SD scores at start compared to patients in the HAG group, all of which may be indicative of poor growth response (35). Moreover, it is also impossible to predict whether higher doses would have resulted in better growth responses.

Futhermore, dosing based on BW resulted in lower but adequate doses for those exhibiting the expected growth response (HAG group). In other words, if dosing based on BSA had been chosen, the majority of patients with HAG (79%) would have received unnecessarily high doses of rhGH, whereas patients in the NHAG group (56%) would have been dosed relatively inadequately low. For patients with the expected height gain in the first year (HAG), subgroup analysis showed that the relative dose difference in favor of BW-based calculations was mostly attributable to the prepubertal group. Among the prepubertal HAG patients, 75% would have been dosed relatively higher if the BSA-based method had been chosen. Similarly, Hughes et al. (7) suggested that BSA-based dosing would result in overtreatment for most children, including those with Turner syndrome and GHD, but not excluding those with obesity. Similar to our findings, both Schrier et al. (8) indicated that younger children would receive more rhGH doses based on BSA in comparison to BW-based doses (11).

These results are important because of the two main drawbacks associated with relative overtreatment with rhGH. Firstly, excessive rhGH may result in potential adverse effects. IGF-1 and IGFBP-3 levels were monitored and were in the reference ranges for our cohort who actually received rhGH doses based on BW while if relatively higher doses had been given using BSA-based methods, there may have been a need for more frequent dose adjustments or clinic visits. Secondly, prescribing higher GH doses to good responders would also result in unnecessary expenditure. Schrier et al. (8) also demonstrated that, despite comparable efficacy, the predicted financial savings for rhGH doses based on BSA and BW would be significantly different. Similarly, the ratio of BSA-to-BW extended to 1.53, indicating that if dosing based on BSA rather than BW had been adopted, the costs would have been 53% higher, despite equal efficacy. Considering that GH treatment is expensive, with an average annual cost of up to 7,088 Euros for a 30-kg child with GH deficit, for children younger than 11 years with BMI levels less than 18, the BSA-based dosage would not be cost-effective in comparison to the BW-based calculations (36).

Study Limitations

Our study was retrospective and we were not able to prescribe both dosings randomly to a larger cohort due to ethical and logistical barriers. Furthermore, due to small group size, we were not able to draw robust conclusions concerning the pubertal group. Another weakness was that the study included two different centers but both centers have been following identical strategies regarding follow-up and rhGH dosing. The greatest strength of our study was that none of the patients had been taking any additional medications, and the safety of treatment was strictly controlled by routinely measured IGF-1/IGFBP-3 values. Thus, the two dosing strategies were hypothetically compared in a homogenous cohort with IGHD who were not overweight and did not have any co-morbidities.

Conclusion

BW- and BSA-based strategies were compared in an homogenous cohort of patients with IGHD receiving rhGH. GH doses based on BW compared to BSA-based dosing may result in the administration of higher doses to children older than 11 years of age with BMI greater than 18 kg/m² and lower doses to children younger than 11 years of age with BMI less than 18 kg/m². Dosing based on BW may be preferable in both prepubertal and pubertal children who do not show adequate growth responses. In prepubertal children, relatively lower doses calculated based on BW rather than BSA provide similar efficacy at lower costs.