SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Aliment Pharmacol Ther 2012; 35: 266–274

Summary

Background  Polycystic liver disease (PLD) is a phenotypical expression of autosomal dominant polycystic kidney disease and isolated polycystic liver disease. Somatostatin analogues, such as lanreotide, reduce polycystic liver volume.

Aim  To establish long-term outcome and safety of lanreotide.

Methods  This was an open-label, observational extension study of a 6-month, randomised, placebo-controlled trial with lanreotide (120 mg/month) in PLD. The length of total treatment was 12 months. Primary endpoint was relative change in liver volume, as determined by CT-volumetry after 12 months of treatment. We offered patients a CT scan 6 months after stopping lanreotide.

Results  A total of 41/54 (76%) patients participated in the extension study. Liver volume decreased by 4% (IQR −8% to −1%) after 12 months of treatment. The greatest effect was observed during the first 6 months of treatment (decrease of 4% (IQR −6% to −1%)). Liver volume remained unchanged during the following 6 months. We found that liver volume increased by 4% (IQR 0–6%) 6 months after end of treatment (n = 22).

Conclusions  Lanreotide reduces liver volume within the first 6 months of treatment and the beneficial effect is maintained in the following 6 months. Stopping results in recurrence of polycystic liver growth. This suggests that continuous use of lanreotide is needed to maintain its effect.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Polycystic liver disease (PLD) is a chronic debilitating disorder with numerous fluid-filled hepatic cysts as its major feature.1 Two inherited conditions are responsible for the majority of PLD cases: isolated polycystic liver disease (PCLD) and autosomal dominant polycystic kidney disease (ADPKD).2 All ADPKD patients have polycystic kidneys, and liver cysts arise in 83% of cases during their lifetime, whereas PLD is the primarily presentation of PCLD.3 Hallmark of PLD is hepatomegaly and livers can increase to eight times their normal size. During the course of PLD, many patients develop symptoms such as pain, early satiety, dyspnoea and restricted mobility.4 Complications such as hepatic venous outflow obstruction and obstructive jaundice occur rarely.5 This disease burden results in a decreased quality of life.4

For decades, PLD has been treated mainly by a variety of surgical procedures aimed to reduce liver volume.6, 7 Unfortunately, the limited effect, mainly because of early recurrence, precludes widespread use.2 Medical treatment options have only become recently available.8 Somatostatin analogues inhibit hepatic cystogenesis and fluid secretion by reducing cAMP in cholangiocytes.9–12 Clinical observations in two polycystic liver patients with ascites suggested that somatostatin analogues reduce liver volume.13 A randomised, placebo-controlled clinical trial with 42 patients showed that octreotide LAR 40 mg reduced polycystic liver volume by 5.9% compared to placebo.14 Similar results were reported in a cross-over trial with 12 patients, where octreotide reduced liver volume by 4.4%, but an increase of 1.2% was observed with placebo.15 In a recent randomised, placebo-controlled, 6-month trial, we showed that lanreotide is superior to placebo for the treatment of PLD in terms of reducing liver volume and improving aspects of quality of life.16 These results suggest that lanreotide is effective for the short-term treatment of PLD, but long-term treatment data are absent. The present study reports the findings of an open-label extension study that followed the 6-month placebo-controlled phase. The primary focus of this 12-month study was to establish the long-term outcome and safety of lanreotide in the treatment of PLD. Furthermore, we want to determine whether a possible beneficial effect is maintained after cessation of lanreotide.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Study population

The eligibility criteria and the study design of the 6-month, randomised, controlled phase of the study have been reported previously.16 Fifty-four patients were enrolled in the study between October 2007 and February 2008. Patients who were >18 years and had >20 liver cysts were eligible for the study. Use of oral contraceptives or oestrogen supplementation was not allowed. The study was performed in Radboud University Nijmegen Medical Centre, Nijmegen, the Netherlands and University Hospital Gasthuisberg, KU Leuven, Belgium.

Study protocol

This was a multicenter open label extension study on safety and efficacy of long-acting lanreotide in adult PLD patients. All patients who had participated in the original trial were invited to participate in an extended treatment protocol to receive additional lanreotide 120 mg deep subcutaneously, every 28 days. To establish a cohort of PLD patients who had had 12 months of active treatment, we allowed all patients who were on placebo in the original trial to start with lanreotide for a total period of 12 months. The patients who received lanreotide in the original trial were treated for an additional 6 months. Patients were offered a CT scan 6 months after end of treatment (Figure 1).

image

Figure 1.  Flowchart extension trial. All patients who participated in the LOCKCYST trial (marked with an asterisk) were asked to continue or start lanreotide for a total of 12 months and 48 of them agreed. The patients who were in the initial placebo group were asked to continue lanreotide treatment after 6 months of treatment, but two of them declined continuation. Finally, 41 patients finished 12 months of lanreotide treatment and 22 received an optitional follow-up CT scan.

Download figure to PowerPoint

Study endpoints

The primary endpoint of this study was relative change of total liver volume from start of lanreotide treatment to 12 months, as determined by CT-volumetry. To this end, we performed a CT-scan at the start of the study and after 12 months of treatment. A CT scan 6 months after cessation of the trial was optional and left to the discretion of the treating physician.

Secondary endpoints were relative change of total liver volume from start of lanreotide treatment to 6 months. In addition, we monitored relative change in total kidney volume in ADPKD patients from start to 12 months as a secondary endpoint.

Additional secondary endpoints were change of total liver volume and total kidney volume from 12 months to 18 months.

Further secondary endpoints were relative change in abdominal symptoms as determined by gastrointestinal symptom (GI) questionnaire,17 and health related quality of life (HRQoL), using the Medical Outcomes Study Form (SF-36).18 The SF-36 comprises nine domains (physical functioning, social functioning, physical role functioning, emotional role functioning, mental health, vitality, bodily pain, change in health perception and general health perception) which can be summarised into a physical and a mental component. Lower scores indicate worse HRQoL. All secondary endpoints were assessed from start of treatment to 12 months. To monitor safety, standard clinical and laboratory tests (plasma concentrations of creatinine, AST, ALT, AP, GGT, bilirubin and albumin) were performed at start of treatment, 1, 3, 6, 9 and 12 months. All patients with a decrease of liver volume after 12 months of active therapy were defined as responders.

CT-volumetry

Volumetry of liver and kidneys was performed using Pinnacle® v8.0d (Philips Electronics NV, Eindhoven, the Netherlands). The CT scans were performed on a multidetector CT scanner [Somatom Sensation 16 (Radboud University Nijmegen Medical Centre) or 64 (University Hospital Gasthuisberg KULeuven), Siemens Medical Solution AG, Erlangen, Germany]. Every patient was scanned by the same CT scanner at each time point. All CT scans were anonymised and dates were removed for volumetry. CT-volumetry was performed by one independent researcher which was unaware of the compound received by the patients. We measured total kidney volume by adding right and left kidney volumes. Liver volumetry was performed following earlier published protocols.16 All individual patients were scanned by the same CT scanner following the same scan protocol. Although the spatial resolution of Somatom Sensation 64 is better than Somatom Sensation 16 this did not affect the quality of the volumetry.

Statistical analysis

Statistical analysis was performed on all patients who finished 12 months of lanreotide. Demographics of these patients were shown in median and interquartile range (IQR). Liver and kidney volumes were assessed at start (T0), after 6 months (T6), and at end (T12) of treatment, and 6 months after stopping lanreotide (T18). The volume data followed a skewed distribution, therefore we present them as median values with interquartile range (IQR). Tables were provided for all questionnaires and laboratory results at start (T0) and end (T12) of lanreotide treatment in mean and standard deviation (SD).

Changes in demographics and in liver and kidney volume between the initial treatment groups were tested by the Mann–Whitney U-test. Changes between the different time points in liver and kidney volume were tested using the Wilcoxon signed rank test. Abdominal symptoms, HRQoL and laboratory values were also assessed at start of treatment and 12 months and changes between these intervals were analysed using the paired t-test. All P-values calculated were two-tailed and the level of significance was set at 0.05.

Ethics

The study protocol was approved by the institutional review boards of the participating institutions according to their national guidelines, and the study was performed according to the Declaration of Helsinki and international standards of good clinical practice. All patients provided written informed consent. This extension trial was registered at clinicaltrials.gov (NCT00771888). CONSORT guidelines were followed for the report of the study.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Demographics

The cohort consisted of 41 patients who all were treated with lanreotide for 12 months. There were more ADPKD (n = 25) than PCLD patients (n = 16). The demographics of all patients in the cohort are displayed in Table 1. A total of 19 patients were originally assigned to placebo, whereas 22 started with lanreotide in the original LOCKCYST trial. All 41 patients were subjected to a CT scan at start, at 6 months and after 12 months. A total of 22/41 patients received a follow-up CT, 6 months after stopping of lanreotide (Figure 1).

Table 1.   Demographics and baseline clinical characteristics
 Total (n = 41)FU-group (n = 22)
  1. ADPKD, autosomal dominant polycystic kidney disease; FU, follow up; PCLD, polycystic liver disease.

  2. Data shown as median (IQR).

Age (years)48.9 (42.9–55.5)51.3 (45.1–57.7)
Gender (male/female)7/345/17
Diagnosis (ADPKD/PCLD)25/1615/7
Centre (Leuven/Nijmegen)20/2120/2
Liver volume (mL)4974 (2982–6299)4375 (2568–5943)
Total kidney volume (mL)953 (508–1662)1156 (735–1751)

Primary endpoint

Change of liver volume from start to 12 months.  At baseline, median liver volumes were 4974 mL (IQR 2982–6299). This decreased to 4711 mL (IQR 2977–6355) after 12 months of lanreotide treatment, corresponding with a median decrease of 4% (IQR −8% to −1%, P < 0.05) (Figure 2). Within this cohort, 32 patients (78%) responded to 12 month of lanreotide treatment (Figure 3). There was no evidence that the changes differed between the groups originally assigned to placebo or lanreotide (data not shown).

image

Figure 2.  Relative change in liver volume during lanreotide and follow-up. This figure shows the relative change of liver volume during lanreotide treatment and after stopping of lanreotide. The lanreotide bar shows a significant change between start and end of 12 months of lanreotide treatment (n = 41). The follow-up bar shows the significant change between end of treatment and 6 months after stop of lanreotide (n = 22). A rebound increase in liver volume is seen. Note that the time paths are different, as lanreotide was given for 12 months, whereas follow-up was only 6 months.

Download figure to PowerPoint

image

Figure 3.  Relative change of liver volume from start to 12 months of lanreotide. This chart depicts the absolute change of liver volume for individual patients and the amount of responders to treatment. Every bar represents one patient. The graphic clearly shows that the decrease in liver volume goes up to 27% in the responders.

Download figure to PowerPoint

Secondary endpoints

Change of liver volume from start to 6 months.  Similar to the data obtained with 12 months of lanreotide treatment, we saw that 6 months of lanreotide treatment induced a 4% (IQR −6% to −1%, P < 0.05) lower liver volume. At 6 months the median liver volume had decreased to 4785 mL (IQR 2820–6100). Thus, the change in total liver volume observed after 6 months of lanreotide was maintained upon further continuation of treatment for another 6 months. The treatment was universally effective as 36 patients (88%) responded to 6 months lanreotide with a decrease of liver volume.

Change of kidney volume from start to 12 months.  We had 25 ADPKD patients with polycystic kidneys who could be analysed for an effect of lanreotide on change of total kidney volume. At baseline, this group had a median total kidney volume of 953 mL (IQR 508–1662). Total kidney volume did not change during 12 months of lanreotide treatment (−1% (IQR −5–2%, P = 0.33) (Figure 4). Indeed, in only 15 (60%) patients 12 months of lanreotide therapy resulted in a lower kidney volume (Figure 5).

image

Figure 4.  Relative change in kidney volume during lanreotide and follow-up. This figure shows the relative change of kidney volume during lanreotide treatment and after stopping of lanreotide. The lanreotide bar shows a not significant change between start and end of lanreotide treatment (n = 41). The follow-up bar shows the significant difference between end of treatment and 6 months after stop of lanreotide and an increase in kidney volume is seen (n = 22). Note that the time paths are different, as lanreotide was given for 12 months, whereas follow-up was only 6 months.

Download figure to PowerPoint

image

Figure 5.  Relative change of kidney volume from start to 12 months of lanreotide. This chart depicts the relative change of kidney volume for the 25 individual autosomal dominant polycystic kidney disease (ADPKD) patients and the amount of responders to treatment. Every bar represents one ADPKD patient. The graphic shows that 15 (60%) ADPKD patients responded to therapy.

Download figure to PowerPoint

Abdominal symptoms and health-related quality of life.  Gastrointestinal symptom and SF-36 questionnaires were completed by all patients during treatment with lanreotide. In the first 6 months we collected data for all 41 patients, whereas data for the complete 12 months lanreotide were available for 31 patients (76%). Main reason for loss to follow-up was that questionnaires were not available for analysis. Analysis of the GI questionnaires showed that the item ‘fasting lower abdominal pain’ scored significantly lower after 12 months compared to baseline (P < 0.05), whereas for other items no statistically significant change was found (Table 2). Similar to the results of the initial trial, the only statistically significant change in the SF-36 after 12 months of lanreotide was improvement of the health perception domain (42–59, P < 0.05). All other domains remained unchanged during the 12 months of lanreotide treatment (Table 3).

Table 2.   Severity of abdominal symptoms and mean VAS score
 T0 (n = 31)T12 (n = 31)
  1. VAS, Visual Analogue Scale.

  2. VAS data shown as mean (SD).

  3. Abdominal symptom severity ≥ 2 on a 7-point adjectival scale ranging from 0 to 6.

  4. P < 0.05 vs. T0.

  5. † VAS, scored on a range of 0–100 (0, no pain; 100, worst pain).

Abdominal pain
 In common11/3112/30
 Postprandial13/3010/30
 Fasting9/304/30*
 Unrelated to defecation8/307/29
Epigastric pain
 In common15/3114/31
 During daytime14/3013/29
 At night/asleep12/3010/30
Heartburn8/305/31
Regurgitation12/317/31
Nausea10/317/31
Vomiting4/313/31
Loss of appetite10/3111/31
Postprandial fullness20/3121/31
Shortness of breath15/3113/31
Abdominal distension18/3114/31
Involuntary weight loss3/313/31
VAS pain score†2321
Table 3.   Health-related quality of life scores (SF-36)
 T0 (n = 31)T12 (n = 31)
  1. Data shown as mean (SD).

  2. P < 0.05 vs. T0.

Physical functioning67 ± 2569 ± 24
Social functioning76 ± 2478 ± 25
Physical role functioning47 ± 4556 ± 44
Emotional role functioning85 ± 2884 ± 35
Mental health75 ± 1577 ± 15
Vitality54 ± 2257 ± 23
Bodily pain22 ± 1222 ± 12
Change in health perception42 ± 2559 ± 24*
General health perception54 ± 2155 ± 21
Physical component summary53 ± 1956 ± 21
Mental component summary69 ± 1770 ± 19

Follow-up

We assembled a new cohort consisting of all 22 patients who received a CT scan 6 months after stopping lanreotide to evaluate the effect of lanreotide after cessation of lanreotide. Basic characteristics were comparable to the primary cohort (Table 1).

Change of liver volume.  Change of liver volume during 12 months of lanreotide showed the same trend as observed in the original group. Liver volume decreased after 12 months of treatment from 4375 mL (IQR 2568–5943) to 4246 mL (IQR 2103–5741). Treatment was stopped at that point and 6 months later the patients were subjected to a CT scan. Here, we observed a rebound growth of liver volume with an increase to 4487 mL (IQR 2101–6091), consistent with a growth of 4% (IQR 0–6%, P < 0.05) (Figure 2).

Change of kidney volume.  The 15 ADPKD patients in this subgroup also received CT kidney volumetry 6 months after end of treatment. Similarly to the primary cohort, median total kidney volume remained constant during 12 months of lanreotide treatment, with an initial volume of 1156 mL (IQR 735–1751) at baseline and 1114 mL (IQR 666–1670) after 12 months of treatment. Kidney growth resumed after stopping of therapy, as kidney volume increased to 1089 mL (IQR 748–1761), which is consistent with an increase of 2% (IQR −1–12%, P < 0.05) (Figure 4).

Safety

Treatment was discontinued in five patients (12%) after a median period of 9 months (range 8–9 months). Reasons for discontinuation were liver transplantation (n = 2), hair loss (n = 1), steatorrhoea (n = 1) and fear of progression of disease (n = 1).

Liver enzyme values remained constant during the observation period of 12 months. Similarly, creatinine values remained unchanged, regardless of initial diagnosis (PCLD or ADPKD). During the first 6 months of treatment albumin increased slightly, but significantly (P < 0.01) and remained increased until the end of 12-month treatment (Table 4).

Table 4.   Laboratory results
 T0 (n = 31)T12 (n = 31)
  1. Data shown as mean (SD).

  2. P < 0.01 vs. T0.

Creatinine (μmol/L)81 ± 3674 ± 22
ASAT (U/L)30 ± 2231 ± 24
ALAT (times ULN)1.1 ± 0.51.2 ± 0.5
AF (times ULN)1.2 ± 0.61.3 ± 0.9
GGT (times ULN)4.0 ± 4.43.8 ± 3.9
Bilirubin (μmol/L)11 ± 711 ± 6
Albumine (g/L)42 ± 443 ± 4*
Glucose (mmol/L)5.1 ± 0.95.3 ± 1.0

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

This article describes the clinical outcomes and safety profiles of patients who received lanreotide therapy for PLD for up to 12 months. The main observation was that 12 months of lanreotide caused a median reduction of liver volume of 4%. The effect seemed to be independent of gender, age, diagnosis, genetic cause of the disease, or initial size of the liver. The major caveat is that this study was underpowered to detect a significant effect for one of these variables. The design of our trial allowed us to study the temporal effects of lanreotide administration on liver and kidney volumes. We found that the reduction of total liver volume was realised in the first 6 months of treatment, and that continuation of therapy for another 6 months maintained the beneficial effect. Subsequently, we assessed the effect of cessation and found that in the cohort of 22 patients studied, liver volumes returned to values seen at start of treatment. As a corollary, there are three possible implications for the concept of somatostatin analogue therapy in PLD. First, our data suggest that continuous treatment is needed to maintain the effect; second, the effect is mainly achieved within the first 6 months of therapy; and third, discontinuation results in recurrence of growth of liver volume to values seen at baseline.

Based on our observations and the data from placebo arms of other randomised clinical trials it is possible to assess the natural course of PLD. We observed that in our original trial liver volume increased with 1.6% after 6 months,16 whereas another trial observed an increase of 0.9% after 12 months.14 A third trial showed an increase of liver volume of 1.2% during 6 months.15 One should realise that the first two trials contained both ADPKD and PCLD patients, whereas the latter included exclusively ADPKD patients. Based on these data from placebo arms, the conclusion is warranted that the growth rate of PLD under lanreotide is delayed which is a clear departure from the ‘normal’ natural course. Notably, when lanreotide is stopped this beneficial effect disappears and liver growth returns to its ‘normal’ course. This finding was also observed previously in a patient who received short-acting octreotide for 90 days; liver volume of this patient returned back to baseline at 3 months of follow-up after an initial reduction of 9.4%.19 The caveat of these observations is that they are uncontrolled by nature.

Upon comparison with more invasive treatment options such as cyst fenestration or hepatic resection, medical therapy seems an elegant alternative. Surgical procedures are associated with a relatively high morbidity, while the safety profile of lanreotide is acceptable.

It remains unknown whether the volume reducing effect of lanreotide is maintained when treatment is extended beyond 12 months, but our data lend to the belief that continuous treatment with lanreotide will continue to suppress growth of polycystic livers. Future clinical trials are needed to address this issue.

The results of this extension study confirm that lanreotide is well tolerated and safe. Side effects were in line with the pharmacological profile of the drug. Most patients experienced steatorrhoea and abdominal cramps after the first injection of lanreotide, but symptoms diminished with prolonged treatment, which is consistent with observations in another trial with lanreotide,20 and only led to discontinuation of therapy in a single patient. During the observation period of 12 months we did not record the occurrence of severe side-effects associated with lanreotide, such as glucose intolerance or development of gallstones.21, 22

Although we do not have head-to-head data available, the present results allow comparison of the relative effects of both octreotide and lanreotide. From the data available from one trial it can be inferred that 12 month treatment with octreotide 40 mg monthly has a slightly larger effect on liver volume relative to lanreotide 120 mg monthly (−4.9% vs. −4%).14 However, during the first 6 months, liver volume decrease does not seem to differ between octreotide and lanreotide (−4.4% vs. −4%).15

One important limitation is that the inclusion criteria of both trials differed. Our trial did include patients on basis of the number of liver cyst (>20) whereas patients in the octreotide trial of Hogan et al.14 needed to have liver volume of >4000 mL. Indeed, the mean liver volumes at baseline differed (5908 mL vs. 4950 mL), and there is some evidence from post hoc analysis that patients with larger liver volume relatively benefit most.16 However, we could not confirm this with the current data set.

The inclusion of ADPKD patients allowed us to study the renal volumes of the polycystic kidneys. Renal volume change was a predetermined secondary outcome measure and hence our study was not powered to draw definite conclusions on this issue. During placebo, renal volume increased by 2.9%, whereas it remained unchanged during treatment with lanreotide. After discontinuation, renal volume increased again by 2%.

As somatostatin analogue treatment is expensive, it is key to separate the responders from the nonresponders. In this trial, we found that 22% of patients did not respond to therapy. Our data suggest that the response rate after 6 months of treatment can be used to differentiate between responders and nonresponders. However, the size of this study does not allow extensive subanalysis to establish a profile of the responders and nonresponders.

Some data suggest that the shortest time to observe a clinical effect of long-acting somatostatin analogues on liver volume is 3 months.19

Finally, in all the studies with lanreotide a dose of 120 mg was used. An ongoing trial currently evaluates the effect of a lower dose (clinical trails.gov identifier NCT01315795).

Our trial comes with some limitations and strengths. One obvious limitation is that the extension phase did not possess the methodological rigour that came with our randomised controlled trial. Both patients and physicians originally blinded became aware of the treatment received, and one might argue that this influences evaluators’ impressions of efficacy. However, the evaluation of the primary outcome measure was done in a blinded manner using anonymised CT scans. Finally, volumetric data 6 months after discontinuation of the trial were only available for 22/41. The CT scan at this time point had not been scheduled in the original protocol and was optitional. Two main strengths of this study are that we established the effect of lanreotide on polycystic liver volume at different time points during treatment and that we were able to evaluate the effect on liver volume 6 months after the last lanreotide dosage.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

This report contributes to the concept that long-acting somatostatin analogues, in particular lanreotide, are both safe and effective in controlling PLD liver volume. We saw that 78% of patients responded to lanreotide with a decrease of liver volume. Our data suggest that somatostatin analogues will be useful in managing patients with (severe) PLD.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

The authors thank Mieke Lamers, Department of Gastroenterology and Hepatology, RUNMC, for comments and contribution to the discussion; Henk Huizenga, Department of Radiation Oncology, RUNMC, for his technical assistance with 3D volumetry; Werner van Steenbergen, Chris Verslype, David Cassiman and Wim Laleman, Department of Hepatology, University Hospital Leuven, for the referral of patients. Declaration of personal interests: None. Declaration of funding interests: This study was funded in part by Ipsen, Boulogne Billancourt, France.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References