Abstract
Objective:
We conducted a prospective trial (NCT01393483) to investigate the utility of serum soluble mesothelin related peptide (SMRP) and tumor mesothelin expression in the management of esophageal adenocarcinoma (ADC).
Summary Background Data:
Clinical management of esophageal ADC is limited by a lack of accurate evaluation of tumor burden, treatment response and disease recurrence. Our retrospective data showed that tumor mesothelin and its serum correlate, SMRP, are overexpressed and associated with poor outcomes in patients with esophageal ADC.
Methods:
Serum SMRP and tumoral mesothelin expression from 101 patients with locally advanced esophageal ADC were analyzed prior to induction chemoradiation (pre-treatment) and at the time of resection (post-treatment), as a biomarker for treatment response, disease recurrence and overall survival (OS).
Results:
Pre- and post-treatment serum SMRP was ≥1 nM in 49% and 53%, and pre- and post-treatment tumor mesothelin expression was >25% in 35% and 46% of patients, respectively. Pre-treatment serum SMRP was not significantly associated with tumor stage (p=0.9), treatment response (radiologic response, p=0.4; pathologic response, p=0.7) or recurrence (p=0.229). Pre-treatment tumor mesothelin expression was associated with OS (HR, 2.08; 95% CI, 1.14–3.79; p=0.017) but had no statistically significant association with recurrence (p=0.9). 3-year OS of patients with pre-treatment tumor mesothelin expression of ≤25% was 78% (95% CI, 68–89%), compared to 49% (95% CI, 35–70%) among those with >25%.
Conclusions:
Pre-treatment tumor mesothelin expression is prognostic of OS for patients with locally advanced esophageal ADC, whereas serum SMRP is not a reliable biomarker for monitoring treatment response or recurrence.
MINI-ABSTRACT
A prospective clinical trial (NCT01393483) to investigate the utility of serum soluble mesothelin related peptide (SMRP) and tumoral mesothelin expression in patients with locally advanced esophageal adenocarcinoma demonstrated that pre-treatment tumor mesothelin expression is prognostic of overall survival, whereas serum SMRP is not a reliable biomarker for monitoring treatment response or recurrence
INTRODUCTION
Multi-modality clinical management of esophageal adenocarcinoma (ADC) is limited by a lack of accurate clinical evaluation of tumor burden, treatment response and early detection of disease recurrence. Imaging alone is not sensitive enough to establish an accurate clinical stage.1 Patients with localized tumor burden who are treated primarily with surgery are often found to have more extensive disease (T3 or N1) at the time of surgery.2 In patients with loco-regionally advanced disease who are treated with multimodality therapy, assessment of treatment response remains inaccurate after induction chemotherapy as well as after completion of chemo-radiation.3 Following curative-intent resection, frequent endoscopic and radiographic imaging follow-up is not sensitive to identify evidence of early recurrence,4, 5 underscoring the need for a marker that can accurately reflect tumor burden, treatment response and recurrence in patients with esophageal ADC.
We investigated mesothelin as a biomarker in the clinical management of patients with esophageal ADC. Mesothelin is a cell surface antigen with low levels of expression limited to the pleura, peritoneum and pericardium.6, 7 We and others have shown that overexpression of mesothelin is observed in a majority of solid tumors, including in esophageal ADC.6, 8 Mesothelin expression is associated with cancer cell proliferation, adhesion and invasion, as well as with cell signaling and metastases in patients with lung ADC, mesothelioma, triple-negative breast cancer and colorectal cancers.9–12 Tumor mesothelin expression and serum soluble mesothelin related peptide (SMRP) have been shown to be associated with tumor burden, aggressiveness and survival in patients with mesothelioma, triple-negative breast cancer and other solid tumors.10, 13
In a retrospective evaluation of normal esophagus, Barrett’s esophagus (BE), and esophageal ADC (n=125), we have shown that normal esophageal mucosa did not express mesothelin.14 High-grade dysplasia in BE specifically expressed mesothelin, whereas BE tissue with low-grade or no dysplasia did not.14 Forty-six percent of esophageal ADC tumors overexpressed mesothelin; esophageal ADC tumors with BE expressed mesothelin more often than those without BE (58% vs. 35%; p=0.01). High mesothelin expression (>50%) was associated with lymph node metastasis and early disease recurrence (46% vs. 11%; p=<0.05). Serum SMRP levels were elevated in 70% of patients with esophageal ADC (mean, 0.89 nmol/L; range, 0.03–3.77 nmol/L), but not in patients with acid reflux, BE or both. This retrospective study was limited by sporadic availability of serum, especially prior to the initiation of multimodality therapy, the lack of any endoscopic pre-treatment tissue, and the uncertain value of post-chemo-radiation surgical tissue.
Supported by our retrospective data, we conducted a prospective trial (NCT01393483) that investigated the utility of tumor mesothelin and serum SMRP as potential biomarkers for the clinical management of patients with esophageal ADC.
METHODS
Study Population
In this study, we prospectively identified patients with locally advanced (T1–4, N1–3) esophageal ADC scheduled to undergo endoscopic biopsy at Memorial Sloan Kettering Cancer Center between April 2011 and August 2017 (NCT01393483). Key inclusion criteria included patients with locally advanced or suspected esophageal ADC who were scheduled to undergo biopsy by endoscopy or endoscopic ultrasound prior to induction chemoradiotherapy and who had baseline surgical tissue available for staining. Exclusion criteria included patients who were less than 18 years of age, previously underwent treatment with chemoradiotherapy, were diagnosed with esophageal squamous cell carcinoma, had a history of other cancers within 3 years of esophageal ADC diagnosis, had metastatic disease, or were medically unfit for endoscopy, induction therapy or esophagectomy (Figure 1). The primary objectives of this study were to determine what proportion of patients with esophageal ADC expressed tissue mesothelin and elevated serum SMRP, and to evaluate whether serum SMRP was associated with clinical stage, treatment response to induction therapy, or disease recurrence. The exploratory objectives were to evaluate tissue mesothelin expression as a biomarker of recurrence, poor response to chemotherapy, and overall survival, and to quantify the correlation between tissue mesothelin and serum SMRP. Memorial Sloan Kettering Cancer Center’s institutional review board (IRB) approved this study (IRB #11–037A), and all patients provided informed written consent. The procedures, including obtaining informed consent, were conducted in accord with the ethical standards of the Helsinki Declaration of 1975.
Figure 1.
CONSORT diagram.
Study Design
Patients were followed for a period of two years following surgical resection. Serum SMRP was measured prior to induction chemoradiation (pre-treatment), at the time of the surgery (post-treatment) (Figure 2A), and every four months for two years. Tissue collected at the time of endoscopic ultrasound with biopsy (pre-treatment) and at the time of surgery (post-treatment) were assessed for tumoral mesothelin expression. Surgical specimens were assessed for pathologic response.
Figure 2.
Expression of serum SMRP and tissue mesothelin in patients with esophageal ADC. (A) Peripheral blood and tissue samples were analyzed prior to and after induction chemoradiation. Following esophagectomy, peripheral blood was analyzed every 4 months for 2 years. (B) Distribution of patients’ serum SMRP concentration (nM) in peripheral blood (n=101) collected prior to (pre-treatment) and after (post-treatment) induction chemoradiation. Each column represents a patient’s serum SMRP concentration (≥1 nM in red, <1 nM in gray). (C) Representative images of biopsy and surgical specimens stained with mesothelin immunohistochemistry and graded based on percentage of staining reactivity: 0 (0–10% reactivity), 1 (11–25% reactivity), 2 (26–75% reactivity), and 3 (>75% reactivity). Tissue mesothelin expression was stratified as positive (>25% reactivity) or negative (≤25% reactivity). (D) Pie chart displaying the number and percentage of patients stratified by tissue mesothelin expression at pre-treatment and post-treatment time points.
Serum SMRP Assessment
Serum SMRP was quantified using double determinant (sandwich) immunoassay (MESOMARK®, Fujirebio Diagnostics, Inc., Malvern, PA, USA), in accordance with the manufacturer’s specifications. The immunoassay had six calibration curve points (0–32 nM). The platers were then spectrophotometrically measured at 450 nM, using a BioTek ELx808 ultra microplate reader (BioTek Instruments Inc., Winooski, VT, USA). Serum SMRP levels were classified as high (≥1 nM) or low (<1 nM).14
Mesothelin Immunohistochemistry
Tissue samples were collected at two timepoints; these were at the time of endoscopic ultrasound with biopsy and at the time of surgery. Specimens were fixed in formalin and embedded in paraffin, then sliced in 4-μM thick sections. Sections were deparaffinized, rehydrated and heated in sodium citrate buffer (pH 6.0) for antigen retrieval. For immunohistochemistry staining, the prepared slides were incubated with monoclonal anti-mesothelin antibody in a 1:90 dilution for 60 minutes (Vector Laboratories, Ltd, UK). Pancreatic adenocarcinoma was used and referenced as the positive control in each study case. Staining was evaluated by a pathologist with expertise in esophagogastric cancer and scored on a scale of 0 (0–10% reactivity), 1 (11–25% reactivity), 2 (26–75% reactivity), or 3 (>75% reactivity). Mesothelin staining was classified as positive (>25% reactivity) or negative (≤25% reactivity) based on standard pathologic guidelines for tissue staining.
Statistical Analysis
Patient clinicopathologic characteristics were summarized as frequency (percentage) and median (interquartile range). Characteristics were compared between pre-treatment tissue mesothelin expression (≤25% vs. >25%) using Wilcoxon rank-sum test for continuous variables and Fisher’s exact test for categorical variables.
The relationship between serum SMRP (continuous) and tissue mesothelin expression (ordinal grade) at the pre-treatment and post-treatment timepoints was tested using Jonckheere-Terpstra test for trend with 1000 permutations. Association between pre-treatment serum SMRP and clinical stage was assessed using the Wilcoxon rank-sum test.
Response to induction was defined as radiologic response (>30% change in SUVmax on PET/CT scan)15 and pathologic response (>60% estimated treatment effect and node negative on the resected surgical specimen).16 The relationship between pre-treatment serum SMRP expression and each definition of response to induction was assessed using Wilcoxon rank-sum test. Similar analysis was conducted to assess change from pre- to post-treatment serum SMRP (decreased vs. increased/ remained unchanged) using Fisher’s exact test.
Cumulative incidence of recurrence (CIR) was calculated from the date of surgery to the date of any recurrence and compared between groups (serum SMRP <1 nM vs. ≥1 nM; tissue mesothelin expression ≤25% vs. >25%) using Gray’s test. Death without recurrence was treated as a competing risk event, and patients were otherwise censored on the date of last follow-up. Relationships between the factors of interest (serum SMRP or tissue mesothelin expression) and recurrence were quantified using multivariable competing risk regression, adjusting for pathologic tumor and lymph node classification, pathologic treatment response, and radiologic treatment response.
Overall survival (OS) was calculated from the date of surgery until the date of death, and patients were otherwise censored on the date of last follow-up. OS was estimated using the Kaplan-Meier method and compared between groups using the log-rank test. The Cox regression model was used to quantify the relationship between groups (tissue mesothelin expression ≤25% vs. >25%). A series of sensitivity analyses for OS was conducted, considering the change in positive and negative status from pre- to post-treatment tissue mesothelin expression. Relationships between the factors of interest (serum SMRP or tissue mesothelin expression) and OS were quantified using multivariable Cox proportional hazards models, adjusting for pathologic tumor and lymph node classification, pathologic treatment response, and radiologic treatment response.
All statistical tests were two-sided, and statistically significance was defined as p<0.05. Analyses were conducted using Stata 15.1 (StataCorp, College Station, TX, USA) and R 4.1.2 (R Foundation for Statistical Computing, Vienna, Austria).
RESULTS
Study Design and Patient Characteristics
Study recruitment initially included three groups of patients with esophageal ADC; the recruitment for two of these groups—patients treated with endoscopic resection (group 1) and patients treated with surgery as their primary therapy (group 2)—was stopped due to slow accrual and thus not included in this analysis. There were 219 patients who underwent induction therapy prior to planned surgical resection enrolled (group 3; Figure 1). Patients who did not meet eligibility criteria were excluded from analysis. These included patients who did not receive induction therapy (n=23) because they were treated at an outside facility (n=12), were lost to follow-up (n=6), developed a second malignancy after enrollment into the trial (n=3), or withdrew from the study (n=2). As specified in the protocol, patients were also excluded from analysis if they received induction chemoradiation but did not proceed with esophagectomy (n=83) because they had progression of disease during induction therapy (n=55), died prior to surgery (n=11), were poor surgical candidates (n=7), demonstrated complete response and did not pursue surgery (n=7), underwent an aborted surgery (n=2), or withdrew from the study (n=1). Patients were also excluded from analysis if baseline tissue was not available for staining (n=12). Fifty-five patients (28% of patients who underwent neoadjuvant chemoradiation) had progression of disease. Among the 101 patients analyzed, 90% were male (n=91), 67% were ever-smokers (n=68) and 79% were classified as clinical stage III (n=80). The median age was 64 years (interquartile range [IQR], 58–70 years). Median duration of follow-up using reverse Kaplan-Meier method was 5.31 years (IQR, 4.10–6.76).
Serum SMRP and Tissue Mesothelin Expression in Patients with Esophageal ADC
There were 49 among 101 patients (49%) who had serum SMRP greater than ≥1 nM at pre-treatment and 49 among 93 patients (53%) at post-treatment (Figure 2B); 8 patients’ post-treatment SMRP values were not available.
35 of 101 patients’ (35%) pre-treatment tissue specimens had mesothelin expression >25% (Figure 2C and 2D). Twenty-seven post-treatment specimens were not able to be evaluated for mesothelin expression due to ≥90% treatment response observed on pathology. Among the 74 patients with post-treatment specimens, 34 (46%) post-treatment specimens had mesothelin expression >25%.
Using Jonckheere-Terpstra test for trend, pre-treatment serum SMRP and tissue mesothelin expression showed positive correlation (p =0.014; Figure, Supplemental Digital Content 1A), while post-treatment serum SMRP showed no statistically significant correlation with tissue mesothelin expression (p=0.49; Figure, Supplemental Digital Content 1B).
Serum SMRP as a Biomarker
Patients classified as clinical stage II (n=5), III (n=80) and IVA (n=16) had a median pre-treatment serum SMRP value of 1.0 nM (IQR, 0.7–1.3 nM), 0.9 nM (IQR, 0.6–1.3 nM), and 1.0 nM (IQR, 0.6–1.3 nM), respectively (p=0.9; Figure 3A).
Figure 3.
Serum SMRP is not significantly associated with clinical stage, treatment response, or disease recurrence. Distribution of patients’ pre-treatment serum SMRP stratified by (A) clinical stage (II, n=5; III, n=80; IVA, n=16) (B) treatment response defined as >30% change in SUVmax on PET/CT (no response, n=21; yes response, n=73), (C) and treatment response defined as >60% treatment response and negative lymph nodes on final pathology (no response, n=56; yes response, n=44). Each boxplot is represented by the median (the line within the box) and interquartile range (the height of the box). (D) Cumulative incidence of recurrence of patients stratified by pre-treatment serum SMRP (<1 nM vs. ≥1 nM), revealing no significant difference (p=0.226). (E) Line plot of the fold change in serum SMRP (SMRPfollow up/SMRPpre-treatment) in patients with recurrence in red (n=40) and those without in gray (n=61) over the duration of study. Date of surgery is represented by time 0 month.
To determine the response to induction therapy, radiologic and pathologic criteria were utilized. Pre-treatment serum SMRP was not significantly associated with radiologic response to induction therapy (p=0.4; Figure 3B). Patients who had radiologic response had a median serum SMRP value of 1.0 nM (IQR, 0.6–1.3 nM) compared to those without a radiologic response who had a median serum SMRP value of 0.8 nM (IQR, 0.6–1.0 nM). The percent change from pre-treatment to post-treatment serum SMRP was similar between patients who responded (percent change=12.5%; IQR, −20.0–60.0%) and those who did not (percent change=0.0%; IQR, −30.0–25.0%; p=0.8; Figure, Supplemental Digital Content 2).
Pre-treatment serum SMRP was not significantly associated with pathologic response to induction therapy (p=0.7; Figure 3C). Patients with pathologic response had a median serum SMRP of 1.0 nM (IQR, 0.5–1.2 nM) versus 0.9 nM (IQR, 0.6–1.3 nM) among those without. The percent change from pre-treatment to post-treatment serum SMRP was similar between patients with response (percent change=12.5%; IQR, −33.3–90.0%) and those without (percent change=9.1%; IQR, −22.2–60.0%; p=1).
When pre-treatment serum SMRP was analyzed as a continuous variable, there was no significant association with disease recurrence (hazard ratio [HR], 0.76; 95% CI, 0.38–1.50; p=0.4). Stratifying pre-treatment serum SMRP as high (≥1 nM) or low (<1 nM) also provided similar observations with recurrence (HR, 0.66; 95% CI, 0.37–1.18; p=0.229; Figure 3D). By examining the longitudinal trend of serum SMRP, we observed that there were no major patterns between those with versus without disease recurrence (Figure 3E).
Tissue Mesothelin Expression and Its Association with Clinical Features, Recurrence, and Survival
We then investigated the utility of tissue mesothelin expression in patients with esophageal ADC who underwent induction chemoradiation followed by esophagectomy. Among 35 patients who had >25% pre-treatment tissue mesothelin expression, 51% also had >25% post-treatment tissue mesothelin expression, compared to 23% with ≤25% expression and 26% not evaluable (p=0.014; Table 1, Figure 4A). Also, patients with pre-treatment tissue mesothelin >25% were of a higher age (p=0.051), with a higher pack-year history (p=0.036). Notably, pre-treatment tissue mesothelin expression was not significantly associated with other clinicopathologic factors such as clinical stage, biopsy grade and lymph node involvement (Table 1; Table, Supplemental Digital Content 1).
Table 1.
Demographics and clinical characteristics of patients
| Pretreatment tissue mesothelin | ||||
|---|---|---|---|---|
| Variable* | Overall (N=101) | ≤25% (N=66; 65%) | >25% (N=35; 35%) | p † |
| Sex | 0.7 | |||
| Female | 10 (10%) | 6 (9.1%) | 4 (11%) | |
| Male | 91 (90%) | 60 (91%) | 31 (89%) | |
| Age, years | 64.1 (58.0, 70.2) | 62.5 (56.5, 69.7) | 67.3 (60.1, 70.8) | 0.051 |
| Smoking status | 0.3 | |||
| Never | 33 (33%) | 25 (38%) | 8 (23%) | |
| Current | 55 (54%) | 34 (52%) | 21 (60%) | |
| Former | 13 (13%) | 7 (11%) | 6 (17%) | |
| Pack years (N=99) | 15.0 (0.0, 34.0) | 8.0 (0.0, 30.0) | 25.0 (2.0, 40.0) | 0.036 |
| cStage (AJCC 8th edition) | 0.3 | |||
| II | 5 (5.0%) | 2 (3.0%) | 3 (8.6%) | |
| III | 80 (79%) | 55 (83%) | 25 (71%) | |
| IVA | 16 (16%) | 9 (14%) | 7 (20%) | |
| Biopsy grade (N=98) | 0.5 | |||
| Well differentiated | 4 (4.1%) | 2 (3.1%) | 2 (5.9%) | |
| Moderately differentiated | 75 (77%) | 51 (80%) | 24 (71%) | |
| Poorly differentiated | 19 (19%) | 11 (17%) | 8 (24%) | |
| Histology | 0.9 | |||
| No residual carcinoma | 18 (18%) | 12 (18%) | 6 (17%) | |
| Minimal residual carcinoma | 20 (20%) | 14 (21%) | 6 (17%) | |
| Invasive carcinoma | 63 (62%) | 40 (61%) | 23 (66%) | |
| Histology grade (N=78) | 1 | |||
| Well differentiated | 2 (2.6%) | 1 (2.0%) | 1 (3.7%) | |
| Moderately differentiated | 57 (73%) | 37 (73%) | 20 (74%) | |
| Poorly differentiated | 19 (24%) | 13 (25%) | 6 (22%) | |
| pStage (AJCC 8th edition; N=98) | 0.8 | |||
| 0 | 16 (16%) | 10 (16%) | 6 (18%) | |
| I | 22 (22%) | 14 (22%) | 8 (24%) | |
| IIA | 6 (6.1%) | 5 (7.8%) | 1 (2.9%) | |
| IIB | 19 (19%) | 13 (20%) | 6 (18%) | |
| IIIA | 5 (5.1%) | 2 (3.1%) | 3 (8.8%) | |
| IIIB | 28 (29%) | 19 (30%) | 9 (26%) | |
| IVA | 2 (2.0%) | 1 (1.6%) | 1 (2.9%) | |
| R resection | 0.12 | |||
| 0 | 97 (96%) | 65 (98%) | 32 (91%) | |
| 1 | 4 (4.0%) | 1 (1.5%) | 3 (8.6%) | |
| Post-treatment tissue mesothelin expression | 0.014 | |||
| ≤25% | 40 (40%) | 32 (48%) | 8 (23%) | |
| >25% | 34 (34%) | 16 (24%) | 18 (51%) | |
| Not available | 27 (27%) | 18 (27%) | 9 (26%) | |
Median (IQR) or frequency (%);
Fisher’s exact test; Wilcoxon rank-sum test; Pearson’s Chi-squared test;
Abbreviations: cStage; clinical stage; AJCC, American Joint Committee on Cancer; pStage, pathologic stage
Figure 4.
Pre-treatment tissue mesothelin expression is associated with overall survival. (A) Alluvial plot revealing differences in the patients’ tissue mesothelin expression at pre-treatment and post-treatment. (B) Kaplan-Meier curve revealing a statistically significant association between pre-treatment tissue mesothelin expression and OS. Survival of patients with pre-treatment tissue mesothelin >25% was worse than that of patients with pre-treatment tissue ≤25% (HR, 2.08; 95% CI, 1.14–3.79; p=0.017).
Pre-treatment tissue mesothelin expression had no statistically significant association with cumulative incidence of recurrence (p= 0.840; Figure, Supplemental Digital Content 3); however, patients with pre-treatment tissue mesothelin expression >25% had a significantly worse OS compared to those with pre-treatment tissue mesothelin expression ≤25% (HR, 2.08; 95% CI, 1.14–3.79; p=0.015; Figure 4B). The 3-year OS among patients with pre-treatment tissue mesothelin expression >25% was 49% (95% CI, 35–70%), compared to 78% (95% CI, 68–89%) among patients with pre-treatment tissue mesothelin expression ≤25% (Figure 4B). A multivariable analysis confirmed pre-treatment tissue mesothelin expression was an independent factor of OS (HR, 2.90; 95% CI, 1.39–6.05; p=0.005; Table 2), after adjusting for pathologic tumor and lymph node classification, pathologic treatment response, and radiologic treatment response.
Table 2.
Multivariable Cox proportional hazards models for overall survival
| Variable | HR | 95% CI | p |
|---|---|---|---|
| Pre-treatment tissue mesothelin expression >25% | 2.90 | 1.39, 6.05 | 0.005 |
| pT 1 or 2 (AJCC 8th edition) | Reference | ||
| pT 0 | 0.49 | 0.13, 1.75 | 0.3 |
| pT 3 or 4 | 1.66 | 0.78, 3.53 | 0.2 |
| pN 1, 2 or 3 vs. pN 0 | 0.90 | 0.34, 2.36 | 0.8 |
| Radiologic response (>30% change in SUVmax) | 0.38 | 0.16, 0.86 | 0.021 |
| Pathologic response (>60% treatment effect with pN 0) | 0.72 | 0.25, 2.04 | 0.5 |
Abbreviations: HR, hazard ratio; CI, confidence interval; pT, pathologic primary tumor status; pN, pathologic nodal status
In contrast, post-treatment tissue mesothelin expression was not significantly associated with OS (p=0.76; Figure, Supplemental Digital Content 4). The 3-year OS among patients with post-treatment tissue mesothelin expression >25% (n=34) was 64% (95% CI, 49–83%) and 63% (95% CI, 49–81%) among patients with post-treatment mesothelin expression ≤25% (n=40).
DISCUSSION
In this prospective study of mesothelin as a biomarker in patients with esophageal ADC, we demonstrated that pre-treatment tissue mesothelin expression was associated with OS, whereas serum SMRP was not associated with tumor burden, treatment response or recurrence. Serial measurement of tissue mesothelin and serum SMRP were not shown to have a trend for monitoring esophageal ADC progression or regression. Despite the above observations showing a lack of association between serum SMRP and tumor burden, lymph node metastasis, or treatment response, pre-treatment tissue mesothelin expression is associated with poor survival. Possible explanations include association of mesothelin expression with tumor invasion17, 18 or triggering other pathways that influence cancer cell aggressiveness.7, 19
Mesothelin is a glycophosphatidylinositol-anchored cell-surface protein that is shed in serum. Serum SMRP was shown to be a marker reflective of tumor burden in patients with malignant pleural mesothelioma epithelioid subtype, where cancer cell mesothelin expression is uniform and intense in >90% of tumor cells.12, 13 However, mesothelin expression in patients with other solid tumors was heterogenous, including in patients with esophageal ADC, which can possibly explain the lack of association of serum SMRP with tumor burden, treatment response and recurrence.8, 11, 20 Similar observations have been published in patients with pancreatic ADC despite high tissue mesothelin expression.21–23 Serum SMRP levels were shown to be influenced by several factors, such as age, weight, blood glucose, lung function tests, creatinine, and blood urea nitrogen;24 we did not adjust the serum SMRP levels for these factors.
In a study published by Moentenich and colleagues, tissue mesothelin expression was not associated with a survival difference.25 It is important to note, however, that analysis was limited to surgical specimens of patients who had previously been treated with chemoradiation or chemotherapy alone. This is comparable to our post-treatment tissue mesothelin cohort, where we also did not detect a survival difference (Figure, Supplemental Digital Content 4). Lack of assessment of tissue mesothelin from multiple sites may have confounded the results. Another plausible reason for the lack of association was that survival outcomes, among other factors, immunologically are a balance between antigen-expressing tumor burden and antigen-specific immune responses. Mesothelin-specific T-cell responses were associated with better outcomes in patients with pancreatic ADC.26, 27 We did not measure mesothelin-specific immune responses in our trial. Lymph node metastases contributes to higher stage in esophageal ADC. We observed that mesothelin was commonly expressed at the invasive edge of patients with esophageal ADC (data not shown); however, it was not associated with lymph node metastases. Pre-treatment tissue mesothelin expression >25% trend towards patients who are of a higher age and have a higher pack-year history, though it does not reach statistical significance; similar observations associating smoking history with tissue mesothelin expression have been made in patients with early-stage lung adenocarcinoma.11
The lack of mesothelin overexpression in normal tissue, overexpression in cancer tissue, and association with cancer aggressiveness and poor outcomes supports mesothelin as a target antigen for antibody, immunotoxin and adoptive cell immunotherapies.28, 29 Despite the lack of support to use tissue mesothelin or serum SMRP as a marker for management, our study has shed light on an eligible cohort for targeting mesothelin in patients with esophageal ADC. In a recently published, phase I clinical trial that investigated the role of mesothelin-targeted chimeric antigen receptor (CAR) T-cell therapy in patients with malignant pleural disease, our group demonstrated safety and feasibility and early evidence of efficacy.28 Consistent with our retrospective observation, confirmation of tissue mesothelin expression in >25% of tumor cells in both pre-treatment and post-treatment of at least one in three patients, and its association with survival, supports investigation of mesothelin-targeted CAR T-cell therapy for patients with esophageal ADC, both locoregional and systemic administration.30 However, heterogenous expression of tissue mesothelin and lack of association with serum SMRP poses a problem in screening patients for eligibility, which is a known hurdle in conducting adoptive T-cell therapy trials for patients with solid tumors.
The prospective nature of investigation—analyzing both serum and tissue mesothelin expression at multiple timepoints—is a strength of our study. Similar retrospective studies have been completed using serum SMRP in patients with malignant pleural mesothelioma;13 however, to our knowledge, this is the first study to do so prospectively in any solid tumor. Forty-two percent of patients did not undergo surgery following neoadjuvant chemoradiation (28% due to progression of disease), which is similar to published literature. A systematic review of 25 randomized clinical trials reported up to 38% of patients who were not offered surgery in an intention-to-treat analysis.31
The limitations of our study include the lack of a cohort of patients treated with surgical resection as their primary treatment. Analysis of this cohort’s tissue mesothelin expression as it relates to overall survival could have provided better context to our observations and given credence to using treatment naïve tissue mesothelin expression as a prognostic factor for survival. However, as noted in our study, this patient population is a smaller proportion of esophageal ADC, and a difficult one in which to conduct biomarker studies. Additionally, we did not find pathologic tumor and lymph node classification as significant prognostic factors in a multivariable analysis, even though these are well accepted determinants of patients’ survival.32 Small sample size could explain this discrepancy, and our results should be interpreted within the context of this limitation. Patients with progression of disease were excluded from the study; serial samples were not collected, as their biomarker results could have potentially been influenced by subsequent therapies.
In summary, we have demonstrated that serum SMRP is a poor biomarker to evaluate tumor burden, treatment response and disease recurrence. However, pre-treatment tissue mesothelin expression is readily obtainable, analyzable and associated with overall survival in patients with esophageal ADC.
Supplementary Material
Correlation between serum SMRP and tissue mesothelin expression. Correlation between tissue mesothelin expression and serum SMRP assessed with Jonckheere-Terpstra test for trend with 1000 permutations using serum and tissue samples at (A) the pre-treatment timepoint (p=0.014) and (B) the post-treatment timepoint (p=0.49). Each boxplot is represented by the median (line within the box) and interquartile range (the height of the box).
Percent change in serum SMRP is not associated with radiologic response. Radiologic response was defined as >30% change in SUVmax on PET/CT scan. Percent change from pre-treatment to post-treatment serum SMRP among patients without treatment response was 0.0% (95% CI, −30.0–25.0%). Percent change among patients with treatment response was 12.5% (95% CI, −20.0–60.0%). There was no association between radiologic treatment response and percent change in serum SMRP (p=0.8).
Pre-treatment tissue mesothelin expression is not associated with cumulative incidence of recurrence. Cumulative incidence of recurrence of patients stratified by pre-treatment tissue mesothelin expression (≤25% vs. >25%) revealed no significant associations (p=0.8).
Post-treatment tissue mesothelin expression is not associated with overall survival. Kaplan-Meier curve revealed no association between post-treatment tissue mesothelin expression and OS (p=0.8).
Acknowledgements
We acknowledge excellent editorial assistance from Summer Koop of the Memorial Sloan Kettering Cancer Center Thoracic Surgery Service.
Conflicts of Interest:
P.S.A. declares research funding from ATARA Biotherapeutics; Scientific Advisory Board Member and Consultant for ATARA Biotherapeutics, Abound Bio, Adjuvant Genomics, Bayer, Carisma Therapeutics, Imugene, ImmPactBio, Johnston & Johnston, Orion pharma, OutpaceBio; Patents, royalties and intellectual property on mesothelin-targeted CAR and other T-cell therapies, which have been licensed to ATARA Biotherapeutics, issued patent method for detection of cancer cells using virus, and pending patent applications on PD-1 dominant negative receptor, wireless pulse-oximetry device, and on an ex vivo malignant pleural effusion culture system. All other authors do not have conflicts of interest to disclose.
Memorial Sloan Kettering Cancer Center has licensed intellectual property related to mesothelin-targeted CARs and T-cell therapies to ATARA Biotherapeutics, and has associated financial interests.
Sources of Funding:
The trial is supported by the National Institute of Health (R21CA164585) and the Esophageal Cancer Foundation Fund. D.R., and J.K.C. are supported, in part, by the National Institutes of Health (T32CA009501-33). P.S.A.’s laboratory work is supported by grants from the National Institutes of Health (P30 CA008748, R01 CA236615-01, R01 CA235667 and U01 CA214195), the U.S. Department of Defense (BC132124, LC160212, CA170630, CA180889, and CA200437), the Batishwa Fellowship, the Cycle for Survival fund, the Comedy vs Cancer Award, the DallePezze Foundation, the Derfner Foundation, the Esophageal Cancer Education Fund, the Geoffrey Beene Foundation, the Memorial Sloan Kettering Technology Development Fund, the Miner Fund for Mesothelioma Research, the Mr. William H. Goodwin and Alice Goodwin, the Commonwealth Foundation for Cancer Research, and the Experimental Therapeutics Center of Memorial Sloan Kettering Cancer Center. P.S.A.’s laboratory receives research support from ATARA Biotherapeutics. The research support did not have any involvement in study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the article for publication.
Footnotes
CRediT Statement
Alexander J. Byun: Data curation; Visualization; Roles/Writing – original draft; Writing – review & editing
Rachel A. Grosser: Data curation; Roles/Writing – original draft; Writing – review & editing
Jennie K. Choe: Data curation; Visualization; Roles/Writing – original draft; Writing – review & editing
Nabil P. Rizk: Conceptualization; Investigation; Methodology; Roles/Writing – original draft; Writing – review & editing
Laura H. Tang: Data curation; Investigation; Methodology; Writing – review & editing
Daniela Molena: Writing – review & editing
Kay See Tan: Formal analysis; Investigation; Roles/Writing – original draft; Writing – review & editing
David Restle: Data curation; Writing – review & editing
Waseem Cheema: Data curation; Visualization; Writing – review & editing
Amy Zhu: Data curation; Visualization; Writing – review & editing
Hans Gerdes: Investigation; Methodology; Writing – review & editing
Arnold J. Markowitz: Investigation; Writing – review & editing
Manjit S. Bains: Writing – review & editing
Valerie W. Rusch: Writing – review & editing
David R. Jones: Writing – review & editing
Prasad S. Adusumilli: Conceptualization; Funding acquisition; Investigation; Methodology; Project Administration; Supervision; Visualization; Roles/Writing – original draft; Writing – review & editing
References
- 1.Young PE, Gentry AB, Acosta RD, et al. Endoscopic ultrasound does not accurately stage early adenocarcinoma or high-grade dysplasia of the esophagus. Clin Gastroenterol Hepatol. 2010; 8(12):1037–1041. [DOI] [PubMed] [Google Scholar]
- 2.Scholer AJ, Uppal A, Chang SC, et al. Inaccurate pretreatment staging can impact survival in early stage esophageal adenocarcinoma. J Surg Oncol. 2020; 122(5):914–922. [DOI] [PubMed] [Google Scholar]
- 3.Eng CW, Fuqua JL 3rd, Grewal R, et al. Evaluation of response to induction chemotherapy in esophageal cancer: is barium esophagography or PET-CT useful? Clin Imaging. 2013; 37(3):468–474. [DOI] [PubMed] [Google Scholar]
- 4.Sudo K, Taketa T, Correa AM, et al. Locoregional failure rate after preoperative chemoradiation of esophageal adenocarcinoma and the outcomes of salvage strategies. J Clin Oncol. 2013; 31(34):4306–4310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lou F, Sima CS, Adusumilli PS, et al. Esophageal cancer recurrence patterns and implications for surveillance. J Thorac Oncol. 2013; 8(12):1558–1562. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Chang K, Pastan I, Willingham MC. Isolation and characterization of a monoclonal antibody, K1, reactive with ovarian cancers and normal mesothelium. Int J Cancer. 1992; 50(3):373–381. [DOI] [PubMed] [Google Scholar]
- 7.Pastan I, Hassan R. Discovery of mesothelin and exploiting it as a target for immunotherapy. Cancer Res. 2014; 74(11):2907–2912. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Morello A, Sadelain M, Adusumilli PS. Mesothelin-targeted CARs: driving T cells to solid tumors. Cancer Discov. 2016; 6(2):133–146. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Thomas A, Chen Y, Steinberg SM, et al. High mesothelin expression in advanced lung adenocarcinoma is associated with KRAS mutations and a poor prognosis. Oncotarget. 2015; 6(13):11694–11703. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Tozbikian G, Brogi E, Kadota K, et al. Mesothelin expression in triple negative breast carcinomas correlates significantly with basal-like phenotype, distant metastases and decreased survival. PLoS One. 2014; 9(12):e114900. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kachala SS, Bograd AJ, Villena-Vargas J, et al. Mesothelin overexpression is a marker of tumor aggressiveness and is associated with reduced recurrence-free and overall survival in early-stage lung adenocarcinoma. Clin Cancer Res. 2014; 20(4):1020–1028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Servais EL, Colovos C, Rodriguez L, et al. Mesothelin overexpression promotes mesothelioma cell invasion and MMP-9 secretion in an orthotopic mouse model and in epithelioid pleural mesothelioma patients. Clin Cancer Res. 2012; 18(9):2478–2489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Burt BM, Lee HS, Lenge De Rosen V, et al. Soluble mesothelin-related peptides to monitor recurrence after resection of pleural mesothelioma. Ann Thorac Surg. 2017; 104(5):1679–1687. [DOI] [PubMed] [Google Scholar]
- 14.Rizk NP, Servais EL, Tang LH, et al. Tissue and serum mesothelin are potential markers of neoplastic progression in Barrett’s associated esophageal adenocarcinoma. Cancer Epidemiol Biomarkers Prev. 2012; 21(3):482–486. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lordick F, Ott K, Krause BJ, et al. PET to assess early metabolic response and to guide treatment of adenocarcinoma of the oesophagogastric junction: the MUNICON phase II trial. Lancet Oncol. 2007; 8(9):797–805. [DOI] [PubMed] [Google Scholar]
- 16.Rizk NP, Venkatraman E, Bains MS, et al. American Joint Committee on Cancer staging system does not accurately predict survival in patients receiving multimodality therapy for esophageal adenocarcinoma. J Clin Oncol. 2007; 25(5):507–512. [DOI] [PubMed] [Google Scholar]
- 17.Suzuki T, Yamagishi Y, Einama T, et al. Membrane mesothelin expression positivity is associated with poor clinical outcome of luminal-type breast cancer. Oncol Lett. 2020; 20(5):193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wang Y, Wang L, Li D, et al. Mesothelin promotes invasion and metastasis in breast cancer cells. J Int Med Res. 2012; 40(6):2109–2116. [DOI] [PubMed] [Google Scholar]
- 19.Uehara N, Matsuoka Y, Tsubura A. Mesothelin promotes anchorage-independent growth and prevents anoikis via extracellular signal-regulated kinase signaling pathway in human breast cancer cells. Mol Cancer Res. 2008; 6(2):186–193. [DOI] [PubMed] [Google Scholar]
- 20.Hanaoka T, Hasegawa K, Kato T, et al. Correlation between tumor mesothelin expression and serum mesothelin in patients with epithelial ovarian carcinoma: a potential noninvasive biomarker for mesothelin-targeted therapy. Mol Diagn Ther. 2017; 21(2):187–198. [DOI] [PubMed] [Google Scholar]
- 21.Hassan R, Remaley AT, Sampson ML, et al. Detection and quantitation of serum mesothelin, a tumor marker for patients with mesothelioma and ovarian cancer. Clin Cancer Res. 2006; 12(2):447–453. [DOI] [PubMed] [Google Scholar]
- 22.Hassan R, Laszik ZG, Lerner M, et al. Mesothelin is overexpressed in pancreaticobiliary adenocarcinomas but not in normal pancreas and chronic pancreatitis. Am J Clin Pathol. 2005; 124(6):838–845. [PubMed] [Google Scholar]
- 23.Zhang X, Yu Y, Peer CJ, et al. Low serum mesothelin in pancreatic cancer patients results from retention of shed mesothelin in the tumor microenvironment. Transl Oncol. 2022; 21:101440. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Park EK, Thomas PS, Creaney J, et al. Factors affecting soluble mesothelin related protein levels in an asbestos-exposed population. Clin Chem Lab Med. 2010; 48(6):869–874. [DOI] [PubMed] [Google Scholar]
- 25.Moentenich V, Comut E, Gebauer F, et al. Mesothelin expression in esophageal adenocarcinoma and squamous cell carcinoma and its possible impact on future treatment strategies. Ther Adv Med Oncol. 2020; 12:1758835920917571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Le DT, Brockstedt DG, Nir-Paz R, et al. A live-attenuated Listeria vaccine (ANZ-100) and a live-attenuated Listeria vaccine expressing mesothelin (CRS-207) for advanced cancers: phase I studies of safety and immune induction. Clin Cancer Res. 2012; 18(3):858–868. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Le DT, Wang-Gillam A, Picozzi V, et al. Safety and survival with GVAX pancreas prime and Listeria Monocytogenes-expressing mesothelin (CRS-207) boost vaccines for metastatic pancreatic cancer. J Clin Oncol. 2015; 33(12):1325–1333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Adusumilli PS, Zauderer MG, Riviere I, et al. A phase I trial of regional mesothelin-targeted CAR T-cell therapy in patients with malignant pleural disease, in combination with the anti-PD-1 agent pembrolizumab. Cancer Discov. 2021; 11(11):2748–2763. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Hassan R, Thomas A, Alewine C, et al. Mesothelin immunotherapy for cancer: ready for prime time. J Clin Oncol. 2016; 34(34):4171–4179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Cherkassky L, Hou Z, Amador-Molina A, et al. Regional CAR T cell therapy: an ignition key for systemic immunity in solid tumors. Cancer Cell. 2022; 40(6):569–574. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Nishikawa G, Banik P, Thawani R, et al. Comparison of neoadjuvant regimens for resectable gastroesophageal junction cancer: a systematic review of randomized clinical trials across three decades. J Gastrointest Oncol. 2022; 13(3):1454–1466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Korst RJ, Rusch VW, Venkatraman E, et al. Proposed revision of the staging classification for esophageal cancer. J Thorac Cardiovasc Surg. 1998; 115(3):660–669; discussion 669–70. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Correlation between serum SMRP and tissue mesothelin expression. Correlation between tissue mesothelin expression and serum SMRP assessed with Jonckheere-Terpstra test for trend with 1000 permutations using serum and tissue samples at (A) the pre-treatment timepoint (p=0.014) and (B) the post-treatment timepoint (p=0.49). Each boxplot is represented by the median (line within the box) and interquartile range (the height of the box).
Percent change in serum SMRP is not associated with radiologic response. Radiologic response was defined as >30% change in SUVmax on PET/CT scan. Percent change from pre-treatment to post-treatment serum SMRP among patients without treatment response was 0.0% (95% CI, −30.0–25.0%). Percent change among patients with treatment response was 12.5% (95% CI, −20.0–60.0%). There was no association between radiologic treatment response and percent change in serum SMRP (p=0.8).
Pre-treatment tissue mesothelin expression is not associated with cumulative incidence of recurrence. Cumulative incidence of recurrence of patients stratified by pre-treatment tissue mesothelin expression (≤25% vs. >25%) revealed no significant associations (p=0.8).
Post-treatment tissue mesothelin expression is not associated with overall survival. Kaplan-Meier curve revealed no association between post-treatment tissue mesothelin expression and OS (p=0.8).