Towards new therapies for bronchopulmonary dysplasia.

Despite advances in neonatology, bronchopulmonary dysplasia (BPD) remains a frequent and important consequence of preterm birth. Innovative therapeutic approaches are needed to improve the respiratory outcome of survivors of preterm birth. The drug discovery pipeline has not been particularly successful for BPD. Currently, budesonide mixed with surfactant, mesenchymal stem cells and IGF1 are promising candidates for further research, but many treatment candidates failed. In this thesis we aimed to search for novel treatment strategies for BPD and improve the research methods necessary to develop new therapies along different phases of the drug development pathway.

In a first set of chapters we aimed to improve the methods for explorative and preclinical research on BPD. Animals models remain the cornerstone of BPD research, however many research results in the past were not translatable to humans. We hypothesize that the lack of prematurity in many models plays a role in this difficult translation. In chapter 1 we demonstrated that prematurity alone, in the absence of hyperoxia, results in altered lung function and structure in preterm rabbit pups, in comparison to term control animals. Focusing more on the effect and mechanisms by which preterm birth disrupts normal lung development, instead of focusing on hyperoxia and mechanical ventilation could advance the search for new therapies. A second reason for the difficult translation of positive findings from animal studies could be insufficient use of measures to avoid bias. A classic read-out in BPD research is the mean linear intercept, a read-out for alveolar structure, which is often obtained in a suboptimal way, prone to bias. In chapter 2 we developed and validated a semi-automatic method for rapid and reproducible assessment of mean linear intercept.

In a second set of chapters we used a previously described preterm rabbit model, in which prematurity in combination with exposure hyperoxia results in a lung functional, structural and vascular phenotype reminiscent of BPD, to search for novel therapeutic strategies. We started with target identification, by transcriptome analysis, in chapter 3. 2217 gene transcripts were significantly dysregulated in the lungs of preterm rabbit pups exposed to hyperoxia for 7 days. We identified clusters of dysregulated genes around functions as inflammation, lung development, vascular development and metabolism of reactive oxygen species. Furthermore we identified novel therapeutic approaches by upstream regulator analysis.

One of these approaches was the use of simvastatin. In chapter 4, we evaluated the efficacy of simvastatin therapy on lung disease in preterm rabbits exposed to hyperoxia. We noticed a beneficial effect on lung function and vascular remodeling, which makes it a promising approach for further research. More knowledge on the exact working mechanism is needed, as we were unable to confirm the gene transcription changes predicted by the upstream regulator analysis. Furthermore, control pups treated with simvastatin depicted an unexpectedly high mortality which warrants caution.

A recent clinical trial showed beneficial effects of a repetitively administered surfactant plus budesonide mixture. The important dysregulation of many inflammatory mediators in the transcriptome analysis suggested that the preterm rabbit model is appropriate to investigate this strategy. In chapter 5, we demonstrated that the preterm rabbit model can be used for studies evaluating the effect of intratracheally administered BPD drugs, something that was previously only possible in preterm lambs and baboons. We demonstrated a good distribution of labeled surfactant or saline after intratracheal injection in spontaneously breathing rabbit pups. In chapter 6 we then assessed the efficacy of a mixture of surfactant plus budesonide. We could show that a single dose of 0,25mg/kg budesonide in surfactant was as (and even more) effective than a daily repetitive dosing regimen. This suggests future clinical trials should assess the efficacy of lower cumulative doses (or a single doses) of budesonide in surfactant.

In final set of chapters, we focused on improving the methods for clinical trials in newborns. Any promising therapy for BPD should undergo evaluation in clinical trials. While a difficult task in any population, clinical trials have additional specific challenges in neonates. Also for the availability of research tools, neonates remain an “orphan population”. While neonates experience relatively more adverse events, there is a lack of tools to standardize safety reporting in this population. In chapter 7 we developed, with the input of broad field of stakeholders, a consensus adverse event severity scale, that can make safety information more comparable across trials and centers.

Finally, clinical trials in neonates are ethically difficult. There is paucity of empirical data on the parental perspectives on neonatal clinical trial participation, which we assessed in chapter 8. We showed, with a well-designed survey, that parents are overall contented with their participation. Almost no parents reported anxiety, stress or guilt when looking back at trial participation. A relevant minority was not aware of typical trial characteristics such as equipoise, the possibility of side effects, the presence of a control group, randomization and blinding. Contentment with follow-up was furthermore low. Research teams should ensure effective communication on trial characteristics and follow-up during consent procedures.

In conclusion, we advanced the ongoing search for novel therapies for BPD in several ways. Along several phases of the drug development process, we developed new methods and new hypotheses to boost future research that will help to improve the respiratory outcome of survivors of extremely preterm birth.