Clinical utility of diagnostic guidelines and putative biomarkers in lymphangioleiomyomatosis

Abstract

Background: Lymphangioleiomyomatosis is a rare disease occurring almost exclusively in women. Diagnosis often requires surgical biopsy and the clinical course varies between patients with no predictors of progression. We evaluated recent diagnostic guidelines, clinical features and serum biomarkers as diagnostic and prognostic tools.

Methods: Serum vascular endothelial growth factor-D (VEGF-D), angiotensin converting enzyme (ACE), matrix metalloproteinases (MMP) -2 and -9, clinical phenotype, thoracic and abdominal computerised tomography, lung function and quality of life were examined in a cohort of 58 patients. 32 healthy female controls had serum biomarkers measured.

Results: Serum VEGF-D, ACE and total MMP-2 levels were elevated in patients. VEGF-D was the strongest discriminator between patients and controls (median = 1174 vs. 332 pg/ml p < 0.0001 with an area under the receiver operating characteristic curve of 0.967, 95% CI 0.93-1.01). Application of European Respiratory Society criteria allowed a definite diagnosis without biopsy in 69%. Adding VEGF-D measurement to ERS criteria further reduced the need for biopsy by 10%. VEGF-D was associated with lymphatic involvement (p = 0.017) but not the presence of angiomyolipomas.

Conclusions: Combining ERS criteria and serum VEGF-D reduces the need for lung biopsy in LAM. VEGF-D was associated with lymphatic disease but not lung function.

Figure 1

Biomarker levels in patients with LAM and controls. (a) Scatter plots comparing patients with definite LAM and controls for: VEGF-D, ACE, MMP-2 and MMP-9. All p values were calculated using Mann–Whitney test and corrected for age. Horizontal lines are the group means. (b) Receiver operating characteristic curves for VEGF-D, ACE, MMP-2 and MMP-9.

Figure 2

Stability of VEGF-D over time. Serum VEGF-D over time for 19 patients with LAM. Each line represents an individual patient.

Figure 3

Analysis of diagnostic utility of VEGF-D.(a) Patients are classified based on whether a definite diagnosis of LAM using ERS criteria can be made without reference to biopsy data. Open circles indicate those patients who underwent lung biopsy all of whom were confirmed to have LAM by histological analysis. Solid line shows the proposed cut off of 800 pg/ml VEGF-D used to support the diagnosis of LAM. (b) Algorithm using diagnostic strategy based upon application of ERS criteria with the inclusion of serum VEGF-D measurement prior to lung biopsy. Figure shows the number from the initial 58 patients in whom a definite diagnosis of LAM can be made at each stage.

Figure 4

Correlation of biomarkers with lung function. VEGF-D levels are not correlated with (a) percent predicted FEV₁ or (b) percent predicted TLCO. (c) serum MMP-2 levels are associated with percent predicted FEV₁ and (d) percent predicted TLCO.

Figure 5

Association of VEGF-D with lymphatic involvement and angiomyolipomas. VEGF-D level is associated with (a) lymphatic involvement but not (b) the presence of angiomyolipomas. (Mann–Whitney test).

Figure 6

Correlation of total St. George’s Respiratory Questionnaire score with percent predicted FEV₁and percent predicted TLCO.