We have shown that markers of ER stress and the UPR are increased in type II AECs isolated from IPF lung, confirming that these pathways are activated in IPF patients. We have also shown that cCK-18 is present in IPF AECs, is generated by activation of the UPR in vitro, and is uniquely elevated in the serum of patients with IPF compared to normal and diseased ILD control subjects. This suggests that cCK-18 may be a biomarker of the UPR and AEC apoptosis and that it could be used to monitor therapies modulating the UPR or AEC apoptosis in IPF patients.
The UPR is a mechanism by which the ER attempts to maintain homeostasis when exposed to proteins that are unfolded or folded incorrectly . Initially, the UPR attempts to restore homeostasis within the ER. Failure to restore homeostasis via the UPR leads to apoptosis . Previous studies have shown that the UPR is activated in the alveolar epithelium of human IPF lung [5, 6]. We build upon these data by directly confirming that the UPR is activated in IPF AECs and that circulating levels of cCK-18, a cleavage product of cytokeratin 18 formed during the UPR, is elevated in IPF patients.
A mechanistically informative biomarker should reflect the activity of important biological pathways. Previous biomarker studies in IPF have focused more on measuring levels of proteins that predict prognosis rather than measuring proteins whose levels are predictive of specific pathologic processes; examples include surfactant protein-A and-D (SP-A and SP-D), [24–26] MMP7, [27, 28] and Krebs von den Lungen 6 antigen (KL-6), [29, 30]. CC-chemokine ligand 18 (CCL18) is an additional biomarker that is increased in IPF serum and bronchoalveolar lavage fluid and may predict progression[31–33]. CCL18 is a marker of alternative macrophage activation and stimulates collagen production in normal lung tissue in vitro; [34, 35] however a clear mechanistic role of CCL18 in IPF has not been proven. Unlike previous biomarkers, this study investigates the use of cCK-18 as a marker of the UPR and apoptosis in IPF patients, two pathologic processes that are activated in IPF patients [4–6]. cCK-18 is generated by caspases activated during apoptosis of epithelial cells,  suggesting it is a marker for AEC apoptosis in patients with IPF. In addition, new data in this manuscript (Figure 2) show that cCK-18 is formed following activation of the UPR in lung epithelial cells. These findings suggest that cCK-18 may be a marker for apoptosis or the UPR in IPF patients.
Circulating cCK-18 does not distinguish between activation of the UPR and apoptosis in epithelial cells. Nevertheless, data in Figure 2B demonstrate that epithelial cells generate cCK-18 following activation of the UPR yet before expression of the apoptosis marker annexin V. This suggests cCK-18 may be generated in cells containing levels of active caspase that are insufficient to activate apoptosis, but sufficient to cleave cytokeratin 18. Because the number of alveolar epithelial cells that are immunoreactive for cCK-18 (Figure 3) or active caspase 3  are far greater than the rare TUNEL positive alveolar epithelial cell found in IPF lungs (data not shown), this is the most likely scenario in IPF lung. Proving this will require future identification of more specific biomarkers of the UPR that can be correlated to cCK-18.
We show that cCK-18 could also be a useful diagnostic biomarker that distinguishes IPF from chronic HP and NSIP, independent of age, gender, smoking, disease severity, and other baseline variables. A recent small study found that cCK-18 may also be elevated in the serum of patients with organizing pneumonia . The clinical importance of this finding is uncertain, given that IPF and organizing pneumonia have unique clinical and radiological features . We used chronic HP and idiopathic fibrotic NSIP as disease controls because the clinical, radiologic, and pathologic features of chronic HP and fibrotic NSIP often have substantial overlap with IPF [15, 16, 18, 19]. Distinguishing IPF from HP and NSIP often requires a surgical lung biopsy, a procedure that carries substantial risk [15, 16]. If confirmed in other cohorts, a serum biomarker, such as cCK-18, that distinguishes IPF from other fibrotic ILDs may reduce the need for lung biopsy.
Serum cCK-18 was not associated with severity or progression of IPF. This conflicts with a previous study that showed cCK-18 correlated with physiologic measures in patients with a variety of ILDs . Several possibilities could explain the lack of association seen our study. First, cCK-18 was measured at a single time point. This single measurement may not be representative of activation of the UPR or apoptosis over the course of an individual’s disease. Second, circulating levels of cCK-18 likely reflect the sum of a complex interplay of physiological processes (i.e. production, clearance, metabolism), each of which may impact the cCK-18 level differently in individual patients. In addition, physiologic progression was measured over a time interval of 6-months after cCK-18 measurement, and cCK-18 level might reflect disease activity and progression on a more limited time scale (e.g. days or weeks). Third, although data were adjusted for several clinical and physiologic variables, other confounders may have been present, including occult conditions, unrelated to IPF, that could cause apoptosis of AECs. cCK-18 was also not elevated in the serum of patients during acute exacerbation of IPF. This may suggest that AEC apoptosis is not a prominent feature of acute exacerbation, however this finding requires further study as there was a trend for higher serum levels of cCK-18 in patients during acute exacerbation of IPF. Finally, we also were unable to detect cCK-18 in BAL fluid in patients with IPF. The reason for this is unknown, but may relate to dilution that occurs during the BAL procedure or to different metabolism of cCK-18 in the alveoli compared to the serum.