Differential Expression of Histamine H₃ Receptor In Healthy, Preneoplastic, and Tumoral Pancreatic Tissue: A Pilot Study

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is one of the leading causes of cancer-related death worldwide, with a mortality rate almost equal to its incidence, highlighting the urgent need for novel biomarkers and therapeutic targets. This pilot study aims to investigate the potential value of histamine H₃ receptor (H₃R) as a prognostic biomarker for this lethal disease. We analyzed the H₃R expression in PDAC using the RNA sequencing data set from The Cancer Genome Atlas (Pan-Cancer Atlas). In addition, H₃R protein levels were evaluated by immunohistochemistry in 27 PDAC samples and compared with adjacent preneoplastic pancreatic tissue of the same patient, and with 10 non-related healthy pancreatic tissues. This preliminary study shows that the H₃R is barely expressed in healthy tissue. Interestingly, H₃R was detected in 96% of PDAC samples, and its expression in tumoral tissue was significantly higher when compared with its expression in preneoplastic tissue, and it was associated with a better prognosis in terms of overall survival. Present findings suggest that H₃R may serve as a potential prognostic biomarker in PDAC. Future research aimed at elucidating the role of H₃R in PDAC biology and its prognostic value in larger patient cohorts is warranted:

Introduction

Despite some advances in the field,^1,2 pancreatic cancer remains a highly lethal disease, showing a 5-year survival rate of less than 10%^1,3,4 and pancreatic ductal adenocarcinoma (PDAC) accounts for 95% of the cases. The lack of symptoms in the early phase of the disease leads to late detection, and most of the patients who reach surgical resection develop recurrence.^5,6 Thus, identification of predictive and prognostic biomarkers and evaluation of novel treatment strategies are necessary.^5,7

Histamine regulates several hallmarks in cancer, including cell proliferation, differentiation, modulation of immune responses, and the response to cancer therapy. In addition, the association between histaminergic receptors’ expression and the clinical outcome was reported in different types of cancer. ⁸ Histamine demonstrates a distinct role through its different receptors; stimulation of histamine H₁ receptor (H₁R) causes an enhanced metastatic potential, opposite to histamine H₂ receptor (H₂R) activation, which inhibits the proliferation of tumoral cells. ⁹ Activation of histamine H₃ receptor (H₃R) increased cell proliferation, an effect that was blocked by an H₃R antagonist. ¹⁰ Furthermore, the treatment with clobenpropit (H₃R antagonist and histamine H₄ receptor [H₄R] partial agonist) improved gemcitabine therapy in pancreatic carcinoma cell lines. ¹¹ To our knowledge, there are no studies identifying the expression and role of H₃R in healthy or tumoral pancreatic exocrine tissues.

This preliminary study found that the H₃R is barely expressed in healthy tissue. Interestingly, H₃R expression in tumoral tissue is significantly higher when compared with its expression in preneoplastic tissue, and it was associated with a better prognosis in terms of overall survival.

Materials and Methods

In Silico Data Analysis

The cBioPortal for Cancer Genomics is an open-access resource for interactive exploration of multidimensional cancer genomics data sets.^12,13 H₃R mRNA expression data and de-identified clinical and survival data were extracted from cBioPortal employing the Cancer Genome Atlas (TCGA) pancreatic cancer (PDAC) PanCancer Atlas data set (n=177) (http://www.cbioportal.org/), access date 01-16-25.

Patient cohorts with high and low H₃R RNA sequencing expression data of tumors according to the median or quartiles were further compared by a Kaplan–Meier survival plot performed using TCGA data available within GEPIA2 (http://gepia2.cancer-pku.cn/#survival). The hazard ratio with 95% confidence intervals and log-rank P-value were calculated (access date 01-16-25). ¹⁴

H₃R expression levels in PDAC and adjacent non-tumor tissues were evaluated using the TCGA database and UALCAN web portal (http://ualcan.path.uab.edu/) (access date 07-08-25).^15,16

Patient Selection

Twenty-seven patients with pancreatic ductal adenocarcinoma (PDAC) that underwent surgery at the British Hospital of Buenos Aires, Argentina, between January 2013 and December 2015 were retrospectively studied using archived paraffin-embedded tumor tissue specimens. The clinical and histopathologic data recorded are described in Appendix Table A1.

Survival data were available for 22 and 21 patients in a period of 12 and 60 months, respectively. During the first year, five patients died, two of them due to pancreatic cancer and three of them because of surgical complications. After a period of 60 months, only one patient was alive. The final cohort used in this study excluded patients with pancreatic neuroendocrine tumors and non-epithelial tumors.

Healthy pancreatic excisional samples were obtained from cadaveric liver donors without pancreatic disease, who underwent surgery for liver transplantation (n=10).

The study was conducted in accordance with the Declaration of Helsinki and approved by the institutional review board of the British Hospital (CRIHB #931).

Histopathological and Immunohistochemical Analyses

Histopathological and immunohistochemical assessments were carried out on formalin-fixed paraffin-embedded tissue sections, which included representative samples of PDAC and adjacent preneoplastic and non-tumoral tissue. All cases had confirmed diagnosis of PDAC by histology. The pathology evaluation of pancreatic specimens included tumor size, Histologic grade (G1: well, G2: moderately or G3: poorly differentiated), lymph nodes involvement, margin status, lymphovascular and perineural Invasion. Histological grading and TNM staging were determined according to the World Health Organization (WHO) and Union for international cancer control (UICC) classifications.^17,18,19

Preneoplastic lesions included low and high-grade dysplastic pancreatic intraepithelial neoplasia (PanIN) defined as microscopic non-invasive neoplasms confined to pancreatic ducts, and intraductal papillary mucinous neoplasms (IPMNs) constituted by grossly visible intraductal epithelial neoplasm of mucin producing cells. ¹⁹

H₃R Immunostaining and Scoring

The expression of H₃R in tumor and preneoplastic tissue was evaluated by immunohistochemical staining as it was previously described. ²⁰ Specimens were incubated with a primary rabbit anti-H₃R polyclonal antibody (1:100, catalog number PA5-50681; Invitrogen, Carlsbad, CA) overnight. Immunoreactivity was detected by using the Peroxidase Vectastain Elite ABC-HRP universal kit, according to the manufacturer’s instructions. Preimmune serum was used for blocking, and to replace the primary antibody to detect non-specific binding of the secondary antibodies (PK-6200; Vector Laboratories, Newark, CA). ²⁰

All specimens were processed following identical and standardized staining procedures. All immunohistochemical sections were evaluated by the first author, a trained pathologist, blinded to clinical data. The H₃R immunoreactivity score was obtained by multiplying the intensity (negative, 0; weak, 1; moderate, 2; and strong, 3) by the percentage of stained cells. H₃R expression was considered “positive” if at least 5% of cell specimens showed membranous and/or granular cytoplasmic staining (regardless of the intensity being weak, moderate or strong) (Fig. 1E). For each case, the entire slide was first examined at low magnification (10× objective) to identify H₃R-positive tumor areas. Ductal epithelial regions free of artifacts were preselected, while fields with necrosis or hemorrhage were excluded. Subsequently, hot spots were evaluated at high power (40× objective), and the highest H₃R staining intensity and its corresponding percentage were recorded for scoring. A minimum of five non-overlapping high-power fields was assessed per slide. To verify the staining specificity, T lymphocytes from lymph nodes and cerebellum tissue served as negative and positive control as we have previously reported. ²⁰ Visualization was performed with an optical microscope Leica ICC50 HD (Wetzlar, Germany).

Figure 1.

Bioinformatic analyses of the expression of H₃R and its immunohistochemical detection in PDAC. Kaplan–Meier plot comparing the (A) overall, and (B) disease free survival curves of patients with high versus low H₃R expression in PDAC tumors. Red line: patients with expression levels above the median; Blue line: patients with expression levels below the median. Upper and lower quartiles are shown as dashed lines. Mantel–Cox (Log-rank test) (http://gepia2.cancer-pku.cn/#survival). (C) Box plots show the mRNA expression levels as transcript per million for pancreatic adenocarcinoma (PAAD) and matched TCGA adjacent non-tumor tissues (normal) data (http://ualcan.path.uab.edu/). (D) H₃R mRNA expression grouped by neoplasm histologic grade (http://www.cbioportal.org/). (E) Representative PDAC samples stained with hematoxylin & eosin (H&E) and H₃R immunostaining with different ranges of H₃R score intensity. All tumors were constituted by atypical invasive duct-like structures with a strong desmoplastic stromal response. In our series, only one case was completely negative for H₃R expression (score 0). Positive granulocytes were used as positive controls (arrows). Membranous and granular cytoplasmic staining of H₃R with scores of 80 (1 × 80), 160 (2 × 80), and 120 (3 × 40) are shown. Scale bar = 20 µm.

Statistical Analyses

Statistical analyses were conducted using GraphPad Prism version 8.01 (San Diego, CA). Mann–Whitney non-parametric test was used to compare average scores. Wilcoxon matched-pairs signed ranks test was used for the statistical analysis of differences in protein expression between tumor and preneoplastic tissue pairs. For determination of the association among different variables, Spearman’s rho correlation coefficients and two-tailed significance were determined. Log-rank test and Gehan–Breslow–Wilcoxon test were performed for Kaplan–Meier survival. All statistical tests were two-sided, and a p<0.05 was considered significant.

Results

In Silico Evaluation of Histamine Receptor Expression in PDAC

From a large PDAC cohort obtained from publicly available data sets (TCGA Pan Cancer Atlas, n=177) Kaplan–Meier survival analysis showed that among histamine receptor subtypes, the H₃R seems to be associated with overall survival in the PDAC patients (Appendix Fig. A1) and increased expression was associated with better survival.

To explore the potential prognostic value of H₃R in PDAC tumors were split based on H₃R median and interquartile expression. H₃R expression was associated with better overall and progression-free survival in PDAC patients (Fig. 1A and B).

TCGA RNA sequencing showed no differences in H₃R mRNA expression in pancreatic tumors compared with histopathologically normal adjacent pancreas (Fig. 1C). It is important to point out the reduced number of TCGA samples available for the evaluation in the latter group. In addition, no significant correlations between H₃R expression and other clinicopathologic parameters, including tumor histological grade (Fig. 1D), TNM stage, or lymph node metastasis were observed (data not shown).

H₃R Protein Expression in Pancreatic Tissue

An immunohistochemistry assay was performed on 27 PDAC from surgical pancreatic specimens. Twenty-six tumors (96.3%) exhibited positive membranous and cytoplasmic immunostaining for H₃R protein with a score ranging between 5 and 210. Only one tumor showed a completely negative expression for this marker (Score 0) (Fig. 1E).

Next, H₃R protein expression was analyzed in preneoplastic lesions (n=17), including pancreatic intraepithelial neoplasias and IPMNs. A positive H₃R immunostaining was observed in 15 preneoplastic lesions (88.2%) with scores ranging from 0 to 140. A higher H₃R expression was detected in PDAC tissue compared with the preneoplastic tissue of the same patient (Fig. 2A and B). Considering that peritumoral pancreas is not completely normal since it frequently exhibits morphological changes such as fibrosis and exocrine atrophy, H₃R protein expression was performed in pancreatic excisional samples obtained from patients without a history of pancreatic disease, who underwent liver surgery that included partial pancreatic resection due to the surgical technique (n=10). Healthy pancreatic architecture was confirmed by histology. H₃R staining remained barely detected in normal pancreas (60%) with lower scores (range: 0–40) (Fig. 2C and D).

Figure 2.

H₃R protein expression in pancreatic healthy, preneoplastic, and PDAC tissue. (A) H₃R immunostaining score in preneoplastic tissue (Preneo, score range: 0–140) and PDAC (score range: 0–210). Wilcoxon matched-pairs signed ranks test, p=0.0188. (B) Higher H₃R expression was observed in neoplastic cells (below) in comparison to preneoplastic tissue (above) of the same patient. H₃R expression was membranous and cytoplasmic. Scale bar=20 µm. (C) H₃R was barely expressed in normal ducts of healthy pancreatic tissue compared with PDAC. Scale bar =20 µm. (D) H₃R immunostaining score in pancreatic healthy tissue (score range: 0–40) and PDAC (score range: 0–210). Mann–Whitney test, p=0.0001. (E) Similar H₃R expression was shown in moderately differentiated PDAC (above) in comparison to poorly differentiated PDAC (below). WD: well differentiated PDAC; MD: moderately differentiated PDAC; PD: poorly differentiated PDAC. Mann–Whitney test, P=NS. Scale bar=20 µm. (F) H₃R expression according to different clinicopathological parameters in PDAC patients. Mann–Whitney’s Test. NS: not significant, IQR: interquartile range. (G) Kaplan–Meier survival curve according to the expression of H₃R (Follow-up: 60 months). The best performing threshold was used as the cutoff between the lower and upper quartiles (Q1, lower quartile) Log-rank and Mantel–Cox test: χ² (Chi-square)=3.747, p=0.0529.

In conclusion, H₃R was significantly overexpressed in the PDAC compared with preneoplastic lesions, and healthy pancreas.

Clinicopathologic Parameters and Prognostic Relevance of H₃R Expression in PDAC

The correlation between the clinicopathological characteristics of PDAC patients and H₃R expression in their tumors was subsequently compared. No significant differences were detected between PDAC H₃R expression and the histopathological grade (Fig. 2E), size, lymphovascular invasion, perineural invasion, or lymph node metastasis (Fig. 2F). The demographic and clinicopathologic features associated with the immunohistochemical scores are presented in Appendix Table A2.

Survival analysis showed that patients with higher H₃R protein expression have increased OS (18.05 months) compared with those with lower H₃R expression (13.30 months) (Fig. 2G).

Discussion

This study provides evidence for the first time of the expression of H₃R in PDAC human samples and its potential association with prognosis. We found that among histamine receptor subtypes, H₃R may have potential as a biomarker associated with patients’ overall survival.

Intraductal proliferative changes precede PDAC, leading to preneoplastic lesions such as pancreatic intraepithelial neoplasia (PanIN), IPMNs, and mucinous cystic neoplasms. ⁵ Results show that the H₃R protein is expressed in a diverse range of membranous and cytoplasmic H₃R immunostaining scores in 96% of PDAC samples. H₃R was increased in PDAC compared with preneoplastic lesions of the same patient, while a reduced expression of H₃R was observed in healthy pancreas suggesting that H₃R expression seems to increase with the severity of pancreatic injury. Similar results were observed in other types of cancer, in which H₃R is overexpressed in tumor tissue when compared with the adjacent histopathologically normal, or non-related healthy tissues. ²¹

This upregulation is likely driven by several mechanisms and pathophysiological factors including chronic inflammation, mast cell infiltration, and altered histamine metabolism, all characteristic features of the pancreatic tumor microenvironment and known inducers of histaminergic signaling in other gastrointestinal cancers.^8,9,22

The TCGA data do not replicate the differential expression observed by immunohistochemistry, in part because the TCGA “normal” group comprises only four tumor-adjacent samples rather than healthy donor pancreas, and bulk RNA-seq is influenced by stromal and immune content. In contrast, immunohistochemistry in healthy tissue provides cell-type and subcellular distribution, confirming the H₃R expression pattern in pancreatic epithelium.

H₃R upregulation in tumors has been associated with enhanced proliferative signaling, modulation of cell cycle regulators, and inhibition of apoptosis in other cancers, such as hepatocellular carcinoma, lung cancer, and breast cancers, through pathways involving cAMP/PKA/CREB and downstream cell cycle proteins. In this context, H₃R antagonists/inverse agonists have shown antitumoral efficacy in different preclinical cancer models, highlighting the window of opportunity for H₃R-targeting compounds drug repurposing. ²¹ In contrast, the selective H₃R agonist Rα-methylhistamine reduced tumor growth in cholangiocarcinoma, although both scenarios involved the inhibition of ERK1/2 signaling. ²¹ Therefore, H₃R stands as a rational molecular target whose pharmacological manipulation might inhibit tumor growth through context-dependent mechanisms. The biological outcome of H₃R modulation may depend on multiple factors including, ligand availability, receptor isoform distribution, and downstream signaling pathways. Through the H₁R and H₃R histamine has demonstrated pro-proliferative and metastatic abilities in the PANC-1 cell line, while showing antitumor activities through the H₂R and H₄R.^9,10 Consistently, H₃R was reported to be expressed in pancreatic beta cells in the islets of Langerhans where it could be involved in cell proliferation. ²³ As far as we know, no in vivo studies have investigated the impact of H₃R pharmacological ligands on pancreatic cancer progression.

Recent data demonstrate the expression of H₃R in numerous types of cancer where it is associated with patient survival and tumor progression (reviewed in Lauretta et al). ²¹ High H₃R expression is correlated with a worse prognosis in lung and prostate cancer, glioblastoma, hepatocellular carcinoma, cholangiocarcinoma, and breast cancer. In contrast, our results show that PDAC patients with higher H₃R expression had significantly better overall survival than those with low levels of H₃R, suggesting that H₃R expression may indicate a less aggressive biological phenotype or reflect adaptive responses or immune control rather than promote tumor progression. In agreement with these results, a positive correlation between H₃R expression with overall survival was previously reported in ovarian cancer. ²⁴ Further studies dissecting H₃R isoforms’ expression, receptor dimerization, and associated signaling pathways in pancreatic cancer are essential to clarify its functional role and therapeutic potential.

This study has several limitations that should be acknowledged. Primarily, the sample size was small and derived from a single institution, which restricts the applicability of the findings to wider patient populations. In addition, some patients either underwent adjuvant treatment at external centers or discontinued therapy, limiting the availability of long-term follow-up data.

In conclusion, the combination of transcriptomic and immunohistochemical findings indicates that H₃R may serve as a novel prognostic marker linked to patient outcomes in PDAC. Furthermore, it may offer a new therapeutic target for this highly aggressive cancer that, in most cases, is unresponsive to conventional treatments. These findings lay the groundwork for future research aimed at elucidating the role of H₃R in PDAC biology and its potential prognostic value in larger patient cohorts.

Appendix

Figure A1.

Kaplan–Meier plotter comparing the overall survival of PDAC patients with high versus low histamine receptor subtypes. Red line: patients with expression levels above the median. Blue line: patients with expression levels below the median. Mantel–Cox (Log-rank test) (http://www.cbioportal.org/).

Table A1.

Clinicopathological Characteristics of the PDAC Patients.

Population	Variables	Patient number n=27	Proportion (%)
Clinical features
Age (years)	Mean/range	70.6 (49–87)
Gender	Female	14	51.9
	Male	13	48.1
Type of surgery	PD	22	81.5
	DP	5	18.5
Pathological features
Size (cm)	Mean/range	3.2 (1.4–7)
Histologic grade	G1, well differentiated	4	14.8
	G2, moderately differentiated	16	59.3
	G3, poorly differentiated	7	25.9
Lymphovascular invasion	No	12	44.4
	Yes	15	55.6
Preneoplastic lesions	No	10	37.1
	Low grade	6	22.2
	High grade	11	40.7
Regional lymph node metastases	No	14	51.9
	Yes	13	48.1
Margin status	Negative	20	70.1
	Positive	7	25.9
Perineural invasion	No	9	33.3
	Yes	18	66.7
pT stage	T1	7	25.9
	T2	15	55.6
	T3	5	18.5
pN stage	N0	14	51.9
	N1	9	33.3
	N2	4	14.8
Histological stage	I	11	40.7
	II	12	44.4
	III	4	14.8

PD: pancreaticoduodenectomy. DP: distal pancreatectomy.

Table A2.

Demographics and Clinicopathologic Features Associated With the Immunohistochemical Score of PDAC Patients.

#	Gender	Age	Histological grade	pT stage	pN stage	PNI	LVI	H3R tumor score	H3R preneo score	5-year survival
1	F	83	PD	T2	N0	Yes	Yes	80	10	Unknown
2	M	60	MD	T2	N1	Yes	Yes	20	NA	No
3	M	71	PD	T3	N0	Yes	Yes	160	NA	No
4	M	72	MD	T2	N0	No	No	0	0	No
5	F	86	MD	T2	N0	Yes	No	20	40	No
6	F	87	MD	T3	N0	No	No	180	NA	No
7	M	79	MD	T3	N0	Yes	No	5	NA	No
8	M	72	MD	T2	N1	Yes	Yes	80	90	No
9	M	72	PD	T1	N0	Yes	No	120	0	No
10	F	60	WD	T2	N0	Yes	No	80	40	No
11	F	71	PD	T2	N1	No	Yes	5	20	No
12	M	79	MD	T1	N0	No	No	120	70	No
13	M	60	PD	T2	N0	Yes	No	20	NA	No
14	F	83	MD	T2	N1	Yes	Yes	10	60	No
15	F	49	MD	T2	N1	Yes	Yes	10	NA	No
16	F	68	MD	T2	N1	Yes	Yes	30	NA	No
17	M	77	MD	T1	N1	Yes	Yes	30	NA	Unknown
18	M	74	WD	T2	N0	No	No	200	120	Yes
19	F	83	MD	T2	N0	No	Yes	160	20	Unknown
20	M	68	WD	T1	N0	No	No	40	60	Unknown
21	M	66	MD	T3	N2	Yes	Yes	140	100	Unknown
22	F	53	WD	T1	N2	No	No	140	140	No
23	F	67	PD	T2	N2	No	Yes	40	NA	No
24	M	57	MD	T1	N0	Yes	No	160	20	No
25	F	75	PD	T2	N1	Yes	Yes	40	NA	Unknown
26	F	72	MD	T1	N1	Yes	Yes	40	20	No
27	F	63	MD	T3	N2	Yes	Yes	210	100	No

Pathological staging was defined according to the ninth edition TNM classification of malignant tumors. ¹⁸

pT categories were based on primary tumor size: pT1 (≤2 cm), pT2 (>2–4 cm), pT3 (>4 cm), and pT4 (involvement of major arteries including the celiac axis, superior mesenteric artery, or common hepatic artery).

Nodal status was classified as pN0 (no regional lymph node metastasis), pN1 (1–3 positive nodes), or pN2 (≥4 positive nodes).

WD: well differentiated PDAC; MD: moderately differentiated PDAC; PD: poorly differentiated PDAC; LVI: lymphovascular invasion; PNI: perineural invasion; Preneo: preneoplasic tissue; NA: not available.