Mesothelin is a frequent target for drug development. It is overexpressed in several malignancies with limited expression in normal mesothelial tissue, which makes mesothelin an ideal candidate to treat certain cancers while potentially avoiding off-target toxicity. Currently, there are no FDA approved therapies for mesothelin to date, but clinical trials are underway to evaluate antibody- and CAR-T- based therapies against mesothelin. 

Our recently published work first started with an observation from the clinical trial investigating anetumab ravtansine, an anti-mesothelin antibody drug conjugate (ADC), with or without pembrolizumab, an immune checkpoint inhibitor to PD-1, in patients with mesothelioma, which is known to overexpress mesothelin. In this study, we saw that patients with high levels of soluble mesothelin who were treated with anteumab ravtansine plus pembrolizumab had worse progression free survival compared to those with lower soluble mesothelin levels. Outcomes for the patients on the control arm who received pembrolizumab monotherapy did not differ based on soluble mesothelin levels [1]. Tumor surface mesothelin is the intended target of this ADC, but surface mesothelin can be released into the systemic circulation as soluble mesothelin by multiple proteases. This led to our hypothesis that soluble mesothelin binds to and sinks anti-mesothelin antibody-based therapies prior to the ADC reaching the intended target on the tumor surface. Similar findings are reported with circulating mesothelin, both soluble mesothelin and mesothelin-positive extracellular vesicles, for other drug mechanisms, such as immunotoxins and CAR-T, respectively [2, 3]. To further explore our suspicion about soluble mesothelin, we added anetumab, the mesothelin antibody component without the chemotherapy payload, to plasma samples with soluble mesothelin, and saw that soluble mesothelin levels were reduced with the addition of anetumab, showing that anetumab binds mesothelin both in the circulation. We also saw that the addition of recombinant mesothelin, a surrogate for soluble mesothelin, to anetumab ravtansine decreased cytotoxicity in mesothelioma cell lines. 

Since soluble mesothelin seems to be a mechanism of resistance to anetumab based therapies, we wanted to see how we could reduce soluble mesothelin levels. Inspired by our prior work with @Jacob Orme, MD PhD et al on soluble PD-L1 in melanoma, we looked to plasma exchange as a novel mechanism to reduce soluble mesothelin levels. We found that plasma exchange can consistently reduce soluble mesothelin levels, which was also seen in the case of soluble PD-L1 [4]. Next, we looked to protease inhibitors as a mechanism to stabilize surface mesothelin. There was a signal of improved stability for surface mesothelin, seen as reduced cell confluence when anteumab ravtansine was combined with recombinant mesothelin and protease inhibitors; however, these findings were not consistently significant. Additionally, the overlapping toxicity with two protease inhibitors plus an ADC in humans is unknown, so other approaches are likely more appropriate for patient care. 

Taken together, our results suggest that soluble mesothelin contributes to treatment resistance for mesothelin antibody-based therapies and plasma exchange can decrease soluble mesothelin levels. These findings highlight the need to better understand the stability of tumor surface proteins and the influence that soluble oncogenic proteins have on patient outcomes. Looking to the future, the next important questions to address are if soluble mesothelin levels can optimally match patients to their ideal treatment strategy, the degree of soluble mesothelin reduction needed for meaningful clinical impact, and other novel mechanisms to clear soluble mesothelin. 

1. Mansfield, A. S. et al. Randomized trial of anetumab ravtansine and pembrolizumab compared to pembrolizumab for mesothelioma. Lung Cancer. 195, 107928. 10.1016/j.lungcan.2024.107928 (2024). 
2. Tumor-derived small extracellular vesicles inhibit the efficacy of CAR T cells against solid tumors. Cancer Res. 83, 2790–2806 (2023).
3. Awuah, P., Bera, T. K., Folivi, M., Chertov, O. & Pastan, I. Reduced shedding of surface mesothelin improves efficacy of mesothelin-targeting recombinant immunotoxins. Mol. Cancer Ther. 15, 1648–1655 (2016).
4. Orme, J. J. et al. Therapeutic plasma exchange clears circulating soluble PD-L1 and PD-L1-positive extracellular vesicles. J. Immunother. Cancer. 8, e001113. 10.1136/jitc-2020-001113 (2020).