Encouraged by reports revealing positive effects in some cases with pulmonary LAM, we decided to start sirolimus therapy in patients with impaired pulmonary function and progressive respiratory disease on an individual basis. Our data support and extend the previous observations suggesting that even in an advanced stage of the disease sirolimus can potentially improve lung function. In addition, we found improvement of submaximal exercise capacity assessed by 6-MWD in some patients. However, due to the fact that a significant number of data is missing and the 6-minute walk test is prone to error without confirmative measurements our findings do not allow to draw firm conclusions regarding a relevant functional benefit of sirolimus over time at this point.
In the absence of an effective alternative treatment, lung transplantation is an accepted therapy for end-stage pulmonary LAM and outcome data are comparable to those achieved for other indications. However, due to overall limited long-term survival, lung transplantation may not be considered as a cure for the rather young cohort of patients affected by LAM [15, 16].
Given the fact that TSC1 and TSC2 proteins regulate signalling through the mTOR pathway and the antiproliferative effects of sirolimus on smooth muscle cell growth, mTOR inhibition has emerged as a promising target for therapeutic interventions in pulmonary LAM .
This notion is supported by the results of Bissler et al., demonstrating a mean increase of FEV1 from baseline of approximately 120 ml after six and twelve months of sirolimus therapy in eleven LAM patients. However, only seven patients had abnormal lung function at the time of enrolment (moderate airflow obstruction in three patients and severe reduction of FEV1 in four patients) in this very important trial and no rate of decline prior to initiation of therapy was reported . The present report confirms and extends these findings, in that we found significant improvement of lung function in a subset of patient with severe airflow obstruction and documented functional deterioration. Our findings are in line with a case report of Taille and co-workers, demonstrating a gain of 570 ml in FEV1 within six months of sirolimus therapy for pulmonary LAM starting from a FEV1 baseline of 32% predicted .
In contrast, interim data from an ongoing trial of sirolimus in Great Britain did not indicate an improvement of pulmonary function in four LAM patients with mild to severe airflow obstruction despite sirolimus therapy for twelve months . In this respect, the accompanying editorial of Paul and Thiele in the New England Journal of Medicine provides a valuable insight into the molecular rationale for sirolimus therapy . Nevertheless, the authors point out that the clinical effects of a pharmacological treatment for somatic mutations of TSC1-TSC2 complexes have to be very variable by nature. However, given the small sample size in our own work and the studies of Bissler et al. and Davies and colleagues, even differences in statistical techniques to assess response to therapy may be a crucial factor contributing to outcome discrepancies [7, 8]. Another possible explanation for the lack of a significant effect of sirolimus may be the fact that in the study of Davies et al., adverse events resulted in all but one patient in periods of dose reduction or cessation . We therefore speculate that in this trial mean exposure to the study drug was not enough to achieve an effect on pulmonary LAM. Unfortunately, significant adverse events including hospitalization due to relevant lower respiratory tract infections or sirolimus-induced pneumonitis have also been frequently observed in our patients and required termination of sirolimus even in the presence of initial functional improvement during therapy. Moreover, the occurrence of pneumothoraces as a potential late complication of sirolimus therapy warrants careful monitoring in future studies. Of note, we speculate that due to the lack of a rigorous assessment, our study might have missed a substantial number of minor adverse events and therefore underestimates the overall negative impact of sirolimus administration. Beyond that, it must kept in mind that discontinuation of the drug within several weeks prior to lung transplantation is mandatory in order to avoid dehiscence of the bronchial anastomosis due to impaired wound healing.
In addition the rate of progression of disease is variable with some patients experiencing a long term course lasting for decades and partially reversible airflow obstruction further complicating outcome assessment. In a large cohort of LAM patients with initially only mild impairment of pulmonary function (mean FEV1 75.2%pred.), the average rate of change in FEV1 was reported to be only -75 ml/year [20, 21]. In contrast, our own small study population demonstrated a projected overall loss of 840 ml per year with an initial FEV1 of 36.0%pred. and severe impairment of diffusing capacity. According to the study of Taveira-DaSilva and colleagues, the most important predictors for further functional decline are initially low FEV1 and severely reduced DLCO. Therefore, it is unlikely that the observed positive functional responses for both FEV1 and FVC readings in our patients are explained by individual variable course of the disease or reversal of airflow obstruction alone .
Moreover, our therapeutic approach is supported by the results of the recently published MILES-trial . McCormack and colleagues demonstrated stabilization and to some extent improvement of lung function parameters in LAM patients with moderate lung impairment. Of note, sirolimus was associated with an acceptable safety profile over a treatment period of twelve months in comparison to placebo. Nevertheless, due to the lack of a predefined loss of FEV1 as inclusion criteria, the subset of patients benefiting most from this medical intervention remains to be established.
However, we have to take into consideration that the use of historic pulmonary function tests in our study increased the risk for a lack of standardisation especially for bronchodilator testing. In addition, we acknowledge that the main reason for referral of LAM patients to our center was evaluation for lung transplantation. So, these limitations might have resulted in a significant selection bias overestimating the rate of actual FEV1 loss and the impact of sirolimus treatment in comparison to the overall LAM population. Moreover, despite the fact that progesterone seems not to be effective in reducing the decline of lung function in LAM patients, we cannot rule out the possibility that withdrawal of progesterone therapy might have influenced the subsequent course in some patients .