Pulmonary Alveolar Microlithiasis

Pulmonary alveolar microlithiasis (PAM) is a rare inherited lung disease caused by inactivating mutations in the sodium phosphate co-transporter, SLC34A2, resulting in the widespread deposition of calcium phosphate crystals in the distal airways and alveoli of the lung. Malpighi, a prominent Italian scientist, rst described the gross pathologic appearance of lungs affected by the disease in 1686 [1]. It was not until 1918 that its histopathologic and radiologic features were carefully detailed by Norwegian physician Harbitz [2]. The Hungarian pathologist Puhr named the disease “microlithiasis alveolaris pulmonum” in 1933 [3] Since that time, over 1000 cases have been reported worldwide, and important insights into the pathogenesis of the disease have emerged [4]. Here, we review epidemiology, pathophysiology, notable clinical features, and management considerations.

C. D. Lampl · F. X. McCormack (*)

Division of Pulmonary, Critical Care, and Sleep Medicine, Department of Internal Medicine, University of Cincinnati, Cincinnati, OH, USA

e-mail: lamplck@ucmail.uc.edu; frank.mccormack@uc.edu

K. A. Wikenheiser-Brokamp

Division of Pathology and Laboratory Medicine, and Perinatal Institute, Division of Pulmonary Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA

Department of Pathology and Laboratory Medicine, University of Cincinnati, Cincinnati, OH, USA

e-mail: Kathryn.wikenheiser-brokamp@cchmc.org

J. C. Woods

Center for Pulmonary Imaging Research, Department of Pediatrics, Division of Pulmonary Medicine, and Department of Radiology, Cincinnati Children’s Hospital Medical Center,

Cincinnati, OH, USA

e-mail: Jason.woods@cchmc.org

J. M. Kofron

The Division of Developmental Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA

e-mail: matthew.kofron@cchmc.org

The overall prevalence of PAM is unknown. Cases have been reported worldwide, but the large majority of subjects have been identi ed in Asia (56.3%) and Europe (27.8%). Patients from Turkey, China, Japan, India, and Italy account for over half of the cases in the literature.

While PAM has been described in all age groups, it has been most commonly discovered in the second or third decade of life, often on chest radiographs obtained for incidental chest complaints or for military or job screening. Mariota noted that 35.8% of PAM patients were diagnosed before the age of 20. Newborns and toddlers have been reported, rarely, including twins who died within 12 h of birth [5]. Although most patients are diagnosed prior to age 50 (88.2%), multiple octogenarians with PAM have been appeared in case reports [6, 7]. PAM has no clear gender or ethnic predilection.

PAM is an autosomal recessive disorder with high penetrance, since it transmits vertically and is associated with consanguinity [8]. In a recent review of 1022 cases, 37.2% of patients had a family history of the disease [4]. The Japanese, Turkish, and Italian cohorts have even greater rates of familial occurrence (43–50%), typically in association with consanguineous marriages [5, 9, 10]. Of interest, the frequency of SLC34A2 mutations in the Japanese general population was estimated to be less than 0.008 [11].

Pathogenesis

Accumulation of microliths in the alveoli is the hallmark of PAM. These calculi are round or ovoid in shape, range from 50 to 5000 μm in diameter and are composed primarily of calcium phosphate with small amounts of calcium carbonate, magnesium, and iron [12–16]. Under the microscope, they can be easily highlighted with von Kossa staining and have a lamellar, “onion-skin” appearance. (Fig. 27.1) At autopsy or transplant, resected lungs are enlarged, heavy,

Fig. 27.2  Pathological evaluation of infant PAM lung, showing orientation of stones along interlobular septa. (a) Gross appearance of lung explant from infant with PAM showing sandy appearance of cut surface, (b) Micro-CT of region of lung shown in a, showing pattern of interlobular and intralobular septal hyperdensity, (c) Second harmonic

generation of brillar collagen (red) and autofuorescence (green) showing microliths lining up along alveolar septa consistent with the pattern in b, (d) Higher power view of microliths revealing lamellar structure

non-collapsible, and non-buoyant [17, 18]. Gross sections have a gritty cut surface. (Fig. 27.2) On a micro-CT image of the lung explant from a 2-year-old PAM infant who was transplanted at our institution, the hyperdense calculi are seen lining up along interlobular septa: a pattern that is con-rmed with second harmonic imaging (Fig. 27.2). Although heterotopic ossi cation in patients with PAM is typically con ned to the lungs, extrapulmonary calci cations have been reported in the pleura [19], diaphragm [20], lumbar sympathetic chain, and testicles. It is not clear whether these calci ed lesions outside lung are truly part of the pathogenesis of PAM or chance occurrences.

Initially, alveolar gas exchange remains intact, but as microliths grow in number and size and ll the alveoli, pulmonary architecture is distorted [21]. As disease progresses, patchy infammation develops, and variable degrees of interstitial brosis can be seen [12]. Intimal and medial thicken-

ing of pulmonary vasculature can also be found in more advanced cases, likely due to pulmonary hypertensive changes related to hypoxia [22, 23].

The genetic basis of PAM was discovered in the mid-­ 2000s. Huqun used high-density homozygosity mapping of three Japanese families to identify a chromosomal segment, 4p15.2, that cosegregated with disease and con rmed that the sodium phosphate transporter within that locus contained protein-truncating mutations in SLC34A2. They also demonstrated that SLC34A2 is highly expressed in alveolar type II cells and postulated a role for the transporter in the export of alveolar phosphate [11]. Corut and colleagues also identi-ed SLC34A2 as a PAM gene in seven patients in Turkey [24]. Since those reports, at least 27 unique mutations have been identi ed in 41 patients [4, 25, 26]. (Table 27.1) Mutations appear to cluster in exon 8 in Chinese cases and exons 7 and 8 in Japanese cases [37]. The heterogeneity in

mutations found thus far is inconsistent with a founder effect, and almost all mutations have been homozygous, suggesting identity by descent. Nonsense mutations resulting in premature protein truncation are most common [24, 32, 37]. The few family studies that have been completed demonstrate 100% penetrance with no apparent correlation between genotype and age at disease onset [24]. That more than one gene is involved in the disease process appears unlikely because mutations in SLC34A2 have been identi ed in almost all cases analyzed [4].

Comprised of 13 exons, SLC34A2 encodes a 2280-­nucleotide mRNA. Its product is Npt2b, a sodium phosphate co-transporter expressed on the alveolar surface of the surfactant-producing type II pneumocytes [38, 39]. Npt2b is also expressed in the gut—where it likely functions as the major transporter for uptake of phosphate under conditions where dietary phosphate intake is limited—as well as the breast, liver, testes, prostate, kidney, pancreas, and ovaries [40, 41]. Other sodium phosphate co-transporters include SLC34 family members Npt2a and Npt2c, which are predominantly expressed in the kidneys, and ubiquitously expressed SLC20 family members Pit1 and Pit2. Mouse lungs have been shown to express Pit1 and Pit2 but not Npt2a or Npt2c [42]. The spectrum of transmembrane phosphate transporters expressed in the human lung has not been well characterized.

In the lung, Npt2b expressed on the alveolar epithelium is thought to resorb the phosphate liberated by the catabolism of surfactant phospholipids by alveolar macrophages [38, 39]. In the absence of functional Npt2b, phosphate likely accumulates in the alveolar lining fuid, binds to free calcium, and eventually forms the lamellated microliths that are characteristic of PAM [42]. Microlith formation probably depends on a favorable milieu created by multiple factors, including but not limited to optimal alveolar lining fuid calcium and phosphate concentrations, pH, and presence of nucleating proteins, lipids, and other small molecules. Surfaces that provide a platform for crystal growth may include phosphate within the polar headgroups of phospholipids that form the surfactant monolayer at the air-liquid interface. Better understanding of these factors may help to predict conditions that promote progression of disease and perhaps to develop strategies to inhibit stone formation and growth.

A recently developed mouse model supports the role of Npt2b in the molecular pathogenesis of PAM. Knockout mice with an epithelial deletion of Npt2b develop a progressive process with diffuse alveolar microlith accumulation,

radiographic opaci cation, restrictive physiology, and pulmonary infammation and brosis, closely mimicking the human disease process [42]. While the serum concentrations of calcium and phosphorus were unchanged when comparing the knockout and wild-type mice, calcium and phosphorus concentrations in alveolar lavage fuids increased roughly tenfold in the Npt2b knockout mice, con rming the central role of Npt2b in alveolar calcium and phosphorus homeostasis. Interestingly, alveolar phosphate levels fuctuate with dietary phosphate levels in PAM mice but are low and unaffected by diet in wild-type animals. The microliths isolated from the mice are composed calcium phosphate salts in proportions that are consistent with hydroxyapatite. Monocyte chemotactic protein-1 (MCP-1) and surfactant protein-D (SP-D) were found to be elevated in both the alveolar lavage and serum of knockout mice as well as in the serum of PAM patients (Fig. 27.3).

Interestingly, the Npt2b knockout animals also developed an unexpected increase in alveolar phospholipids. Although this nding has not been reported in humans, to our knowledge, elevated serum surfactant levels [43], oil red O-positive alveolar macrophages [44], and the abundant eosinophilic material reported to ll the alveoli of infant identical twins in PAM [45] may all be consistent with variable degrees of phospholipidosis. Like the mice, these ndings may prove to be related to altered surfactant catabolism by dysfunctional alveolar macrophages.

Microliths transferred into the lungs of wild-type mice produced marked macrophage-predominant infammation and elevation of serum MCP-1 that peaked after 1 week and resolved at 1 month, suggesting that the normal lung has the capacity to dissolve the stones and return to a normal state. EDTA lavage of the lung, ex vivo, was effective at reducing the burden of stones. Administration of a very low phosphate diet to young knockout animals prevented microlith formation and reduced serum SP-D (Fig. 27.3). In older animals with established PAM lesions, low phosphate diet reduced the profusion of hyperdense in ltrates on radiographs and micro-CTs. Mechanisms involved with the ben- e cial effects of low phosphate diet on microlith burden remain unclear, but given preliminary ndings that dietary phosphate intake has a direct effect on alveolar phosphate levels and osteoprotegerin levels in the alveolar lining fuid, possibilities being considered include upregulation of alternative phosphate transporters and activation of pulmonary osteoclast-like cells (unpublished data, personal observations).