Kita-Kyushu lung cancer antigen-1 expression in initial gastric tumors predicts multiple cancer development: A pilot study

Abstract

BACKGROUND

We have previously reported that the cancer/testis antigen Kita-Kyushu lung cancer antigen-1 (KK-LC-1) is frequently expressed in gastric cancer (GC), with a particularly high positivity rate observed in patients with multiple GCs. Thus, we conducted a preliminary study to further investigate the relationship between KK-LC-1 expression and multiple GCs. Specifically, we aimed to explore the potential clinical utility of KK-LC-1 as a biomarker for identifying patients at risk of developing multiple gastric cancers, with a focus on its expression in the initial tumor lesions.

AIM

To investigate the association between KK-LC-1 expression and the development of multiple GCs in a preliminary cohort.

METHODS

Among 124 patients (177 lesions) treated for GC at Toho University Medical Center Ohashi Hospital between September 2020 and November 2023, 109 (single cancer: 82; multiple cancers: 27) with available initial tumor specimens were enrolled. KK-LC-1 expression was assessed using immunohistochemistry, and clinicopathological correlations were analyzed using Fisher’s exact test. Tumor locations were classified anatomically and analyses were performed, focusing on the initial cancer sites.

RESULTS

In tumors located in the upper stomach, KK-LC-1 expression did not correlate with clinicopathological features. However, in the middle and lower stomach, KK-LC-1 expression was significantly associated with older age, differentiated histological types, and multiple cancers. Notably, all patients with multiple GCs had KK-LC-1-positive tumors (100%) in their initial lesions. The KK-LC-1 positivity rate in the initial tumors of multiple cancers in the middle and lower region was significantly higher than that in solitary cancers. Conversely, the absence of KK-LC-1 expression in a solitary tumor located in the middle and lower region may suggest a reduced risk for the subsequent development of multiple primary cancers.

CONCLUSION

KK-LC-1 expression in the initial gastric tumors, particularly in the middle and lower stomach regions, may serve as a predictive biomarker for the future development of multiple GCs.

Key Words: Gastric cancer; Kita-Kyushu lung cancer antigen-1; Multiple cancers; Prognostic marker; Tumor differentiation

Core Tip: We have previously reported that the cancer/testis antigen Kita-Kyushu lung cancer antigen-1 (KK-LC-1) is frequently expressed in gastric cancer (GC), with high expression observed in multiple GCs. Therefore, as a preliminary study, we investigated the relationship between KK-LC-1 expression and multiple GCs, focusing on its potential clinical utility. In the middle and lower stomach regions, KK-LC-1 expression was significantly associated with older age, differentiated histological type, and the presence of multiple cancers, with all cases in the multiple cancer group exhibiting KK-LC-1 positivity (100%). KK-LC-1 expression in primary gastric tumors suggests a higher risk of subsequent development of multiple GCs.

INTRODUCTION

Gastric cancer (GC) is the fourth leading cause of cancer-related death worldwide and ranks fifth in incidence among all cancers[1]. Endoscopic resection is widely used as an effective and minimally invasive local treatment for early GC with a low risk of lymph node metastasis. However, as the entire stomach is preserved following endoscopic resection, the risk of metachronous GC remains, and its incidence is higher compared to that of surgical resection[2,3]. The reported incidence of postoperative metachronous multiple GC is 2.35% following distal gastrectomy, 3.01% following pylorus-preserving gastrectomy, 6.28% following proximal gastrectomy, and 8.21% following function-preserving gastrectomy[4]. In contrast, the incidence after endoscopic treatment ranges from 9.5% to 14.0%[5].

Cancer/testis antigens are a group of tumor-associated antigens expressed in various cancer tissue and in testicular germ cells[6]. Thus, cancer/testis antigens are promising targets for cancer-specific immunotherapy. Among them, Kita-Kyushu lung cancer antigen-1 (KK-LC-1), a cancer/testis antigen with high expression levels in triple-negative breast cancer and GC, has garnered significant attention[7,8]. Currently, various therapeutic strategies targeting KK-LC-1 are under investigation, including T cell receptor-engineered T cell therapy, peptide-drug conjugate therapy, and molecular targeted therapy using small-molecule compounds[9-11].

In our previous studies, we confirmed that KK-LC-1 is a promising therapeutic target, with high expression observed in GC (81.6%) and triple-negative breast cancer (100%)[8,12]. Moreover, we reported that Helicobacter pylori (H. pylori) infection induces KK-LC-1 expression and that the antigen is highly expressed in early-stage GC (79.5%)[13]. Furthermore, KK-LC-1 expression has been detected in gastric mucosa during the precancerous stage[13], and we previously found a high rate of KK-LC-1 positivity among patients with multiple GCs[14]. Even in cases of successful H. pylori eradication, KK-LC-1 expression remained high, suggesting the involvement of pathways independent of H. pylori infection[14]. Therefore, we conducted a preliminary study to further explore the relationship between KK-LC-1 expression and multiple GCs. Specifically, we examined the clinical utility of KK-LC-1 as a biomarker for identifying patients at increased risk of developing multiple GCs, focusing on its expression in initial tumor lesions.

MATERIALS AND METHODS

The Human Ethics Review Committee of the Toho University Ohashi Medical Center, Japan (Approval No. H23034_H20023) approved the study protocol, and all experiments were conducted in accordance with the relevant guidelines and regulations. All patients signed an informed consent form before the tissue samples were collected for this study.

Patients

From September 2020 to November 2023, 124 patients (177 lesions) were treated for GC at the Department of Surgery or Department of Gastroenterology at Toho University Medical Center Ohashi Hospital (Tokyo, Japan). Among them, 39 patients had multiple cancers, accounting for 77 lesions (50 treated endoscopically and 27 treated surgically). Tumor tissues from resected GC were collected and used in this study. Clinicopathological findings were classified according to the Japanese Classification of Gastric Carcinoma (14^th edition)[15]. In addition, to investigate the characteristics of the first cancer lesions, 109 cases (single cancer, 82 cases; multiple cancers, 27 cases) were included in the study, wherein the initial lesion could be evaluated.

Among multiple cancers, synchronous cancers were defined as multiple GCs with a treatment interval of ≤ 1 year, whereas those with a treatment interval > 1 year were defined as metachronous. In this study, papillary and tubular adenocarcinomas were reclassified as differentiated tumors based on histological findings, whereas poorly differentiated adenocarcinoma, signet ring cell adenocarcinoma, and mucinous carcinoma were categorized as undifferentiated tumors. The tumor sites of GC were reclassified as upper, representing the upper third of the stomach, and middle and lower, representing the middle and lower thirds of the stomach. H. pylori infection was evaluated using blood anti-H. pylori antibody (DENKA Kit; DENKA Seikenn Corporation, Tokyo, Japan). A history of H. pylori eradication was confirmed through interviews. H. pylori infection was defined as uninfected when the blood anti-H. pylori antibody was negative, current infection when the blood anti-H. pylori antibody was positive, and previous infection when the blood anti-H. pylori antibody was negative with a history of eradication. Atrophic gastritis was assessed by upper gastrointestinal endoscopy to evaluate the degree of atrophy. The degree of gastric mucosal atrophy was categorized as none, closed type (C-1, C-2, C-3), and open type (O-1, O-2, O-3) according to the endoscopic atrophic-border scale described by Kimura and Takemoto[16].

Immunohistochemical staining

Formalin-fixed, paraffin-embedded tissue sections (3 μm thick) of human GC tumors and adjacent non-tumor areas were prepared for each patient. Antigen retrieval was performed by autoclaving the sections at 97 °C for 20 minutes in a low-pH citrate buffer (pH 6.0). Endogenous peroxidase activity was blocked using 6% hydrogen peroxide. For KK-LC-1 immunostaining, the supernatant of the previously described hybridoma clone 34B3[12], diluted 1:80, was used as the primary antibody and incubated for 1 hour. Staining was conducted using the Envision FLEX staining kit (Agilent Technologies, Santa Clara, CA, United States) on a DAKO Autostainer Link 48 platform (Agilent Technologies, Santa Clara, CA, United States), following the manufacturer’s protocol.

KK-LC-1 expression was evaluated in both tumor and adjacent non-tumor tissues within the same section. In tumor tissue, KK-LC-1 positivity was defined by the presence of stained tumor cells. Staining intensity was scored based on cytoplasmic expression as follows: 0 (negative), 1+ (weak), 2+ (moderate), and 3+ (strong). Specimens were considered KK-LC-1 positive if ≥ 10% of tumor or non-tumor cells exhibited 2+ or 3+ cytoplasmic staining intensity (Figure 1). The immunohistochemical staining of KK-LC-1 was independently evaluated by three gastrointestinal surgeons (Futawatari N, Akimoto Y, and Maehara J) who have extensive experience in histopathological interpretation of GC specimens. After independent scoring, all samples were reviewed with additional input from two board-certified pathologists (Yokouchi Y, and Takahashi K) to confirm the consistency and accuracy of the evaluation.

Figure 1 Immunohistochemical staining of Kita-Kyushu lung cancer antigen-1 with Kmab34B3 of tumor cells in gastric cancer. A: Negative staining; B: Negative staining. Magnification: 100 ×.

Statistical analysis

Comparisons between KK-LC-1 positive and negative groups were performed using Fisher’s exact test for categorical variables and Mann-Whitney U test for continuous variables. All P-values < 0.05 were considered significant. All statistical analyses were performed using EZR version 1.63 (Saitama Medical Center, Jichi Medical University, Saitama, Japan)[17].

RESULTS

Patient characteristics

Table 1 summarizes the characteristics of 39 patients with multiple GCs. The mean age of the patients was 76.4 ± 7.50 years, and most were male (31 male vs 8 female). The numbers of cases with multiple cancers involving two, three, four, and five lesions were 28, 7, 1, and 3, respectively. At least one lesion was positive for KK-LC-1 expression in 94.9% of cases.

Table 1 Characteristics of 39 patients with multiple gastric cancer, mean ± SD/median (interquartile range).

Category
Gender (male/female)	31/8
Age	76.4 ± 7.503
Number (2/3/4/5)	28/7/1/3
Synchronous/metachronous	28/11
Interval (month)
Synchronous	1.9 (0-10)
Metachronous	30 (14-95)
Treatment (ESD/surgery)	26/1
Atrophic gastritis (non/closed/open), n = 37	1/4/32
H. pylori eradication	41.7% (15/36)
Blood anti-H. pylori antibody	32.5% (12/37)
KK-LC-1 expression (more than 1 lesion)	94.9% (37/39)

ESD: Endoscopic submucosal dissection; H. pylori: Helicobacter pylori; KK-LC-1: Kita-Kyushu lung cancer antigen-1.

KK-LC-1 expression in multiple cancer

Details of KK-LC-1 expression in the 39 patients are shown in Figure 2. KK-LC-1 expression was investigated in 67 lesions, of which 53 (79.1%) were KK-LC-1 positive. Only two patients (patient 10 and patient 25) did not show KK-LC-1 expression in either lesion (Table 2). Neither patient was infected with H. pylori, and the initial lesions occurred in the upper region.

Figure 2 The expression of Kita-Kyushu lung cancer antigen-1 was evaluated in the tumor areas of each gastric cancer patient. Open squares indicate negative results, closed squares indicate positive results, and an X indicates cases that could not be tested. The tumor areas were divided into three parts: Upper, middle, and lower. U: Upper; L: Low; M: Middle.

Table 2 Details of two cases in which both lesions were negative for Kita-Kyushu lung cancer antigen-1 expression.

	Lesion	Age	Gender	Interval	H. pylori infection	GA	Histological finding	T	Location	Size (mm)	Treatment
No. 10	1st	81	Male	0	Negative	O-3	por2, tub2	3	Upper	55	Surgery
	2nd						tub1	1a	Middle	20	ESD
No. 25	1st	73	Male	16	Negative	None	tub2	1b1	Upper	10	ESD
	2nd						tub2, tub1	1b1	Middle	11	ESD

H. pylori: Helicobacter pylori; GA: Gastric atrophy; ESD: Endoscopic submucosal dissection.

Correlation between KK-LC-1 expression and clinicopathological factors in initial GC

Table 3 presents the correlation between KK-LC-1 expression and clinicopathological factors in initial GC. KK-LC-1 expression was significantly associated with differentiated histological type (P = 0.035), multiple cancers (P = 0.016), and a history of H. pylori eradication (P = 0.027).

Table 3 Correlation of Kita-Kyushu lung cancer antigen-1 expression and clinicopathological factors in initial gastric cancer, n (%)/median (interquartile range).

	KK-LC-1 expression
	Positive	Negative	P value
Sex			0.341
Male	54 (67.5)	26 (43.3)
Female	23 (79.3)	6 (20.7)
Age	77 (70.0-81.0)	74 (59.8-79.5)	0.078
Depth of invasion			0.110
T1	61 (74.4)	21 (25.6)
T2-T4	16 (59.3)	11 (40.7)
Histological type			0.035
Differentiated	60 (76.9)	18 (23.1)
Undifferentiated	17 (54.8)	14 (45.2)
Tumor diameter			1
≤ 20 mm	38 (70.4)	16 (29.6)
> 20 mm	39 (70.9)	16 (29.1)
Lymph node metastasis			0.073
Positive	12 (54.5)	10 (45.5)
Negative	65 (74.7)	22 (25.3)
Lymphatic invasion			0.371
Positive	23 (63.9)	13 (36.1)
Negative	54 (74.0)	19 (26.0)
Venous invasion			0.254
Positive	20 (62.5)	12 (37.5)
Negative	57 (74.0)	20 (26.0)
Pathological stage			0.150
I	61 (74.4)	21 (25.6)
II, III, IV	16 (59.3)	11 (40.7)
Treatment			0.091
Endoscopic	46 (78.0)	13 (22.0)
Surgery	31 (62.0)	19 (38.0)
Atrophic gastritis (n = 106)			0.355
Yes	72 (72.0)	28 (28.0)
No	3 (50.0)	3 (50.0)
Anti-H. pylori antibody titer (n = 101)			0.279
Positive	27 (64.3)	15 (35.7)
Negative	44 (74.6)	15 (25.4)
H. pylori eradication (n = 99)			0.027
Yes (n = 40)	33 (82.5)	7 (17.5)
No (n = 59)	36 (61.0)	23 (39.0)
Multiple cancers			0.016
Multiple	24 (88.9)	3 (11.1)
Single	53 (64.6)	29 (35.4)
Location			0.800
Upper	17 (73.9)	6 (26.1)
Middle and lower	60 (69.8)	26 (30.2)

H. pylori: Helicobacter pylori; KK-LC-1: Kita-Kyushu lung cancer antigen-1.

Correlation between KK-LC-1 expression and clinical pathological factors in initial GC in the middle and lower region

Table 4 summarizes the correlation between KK-LC-1 expression and clinical pathological factors in initial GC (86 cases) in the middle and lower region. KK-LC-1 expression was significantly higher in patients with older age (P = 0.020), differentiated type (P = 0.022), and multiple cancers (P = 0.002). KK-LC-1 expression was observed in all cases of multiple cancers.

Table 4 Correlation between Kita-Kyushu lung cancer antigen-1 expression and clinical pathological factors in initial gastric cancer in the middle and lower region, n (%)/median (interquartile range).

	KK-LC-1 expression
	Positive	Negative	P value
Sex			0.322
Male	39 (66.1)	20 (33.9)
Female	21 (77.8)	6 (22.2)
Age	77.0 (70.0-81.0)	71.5 (58.3-77.8)	0.020
Depth of invasion			0.192
T1	46 (74.2)	16 (25.8)
T2-T4	14 (58.3)	10 (41.7)
Histological type			0.022
Differentiated	46 (80.0)	13(22.0)
Undifferentiated	14 (51.9)	13 (48.1)
Tumor diameter			0.816
≤ 20 mm	30 (68.2)	14 (31.8)
> 20 mm	30 (71.4)	12 (28.6)
Lymph node metastasis			0.158
Positive	10 (55.6)	8 (44.4)
Negative	50 (73.5)	18 (26.5)
Lymphatic invasion			0.614
Positive	17 (65.4)	9 (34.6)
Negative	43 (71.7)	17 (28.3)
Venous invasion			0.453
Positive	16 (64.0)	9 (36.0)
Negative	44 (72.1)	17 (27.9)
Pathological stage			0.417
I	47 (72.3)	18 (27.7)
II, III, IV	13 (61.9)	8 (38.1)
Treatment			0.097
Endoscopic	38 (77.6)	11 (22.4)
Surgery	22 (59.5)	15 (40.5)
Atrophic gastritis (n = 84)			0.579
Yes	57 (71.3)	23 (28.8)
No	2 (50.0)	2 (50.0)
Anti-H. pylori antibody titer (n = 79)			0.214
Positive	20 (60.6)	13 (39.4)
Negative	35 (76.1)	11 (23.9)
H. pylori eradication (n = 78)			0.142
Yes	26 (78.8)	7 (21.2)
No	28 (62.2)	17 (37.8)
Multiple cancers			0.002
Multiple	16 (100)	0 (0)
Single	44 (62.9)	26 (37.1)

H. pylori: Helicobacter pylori; KK-LC-1: Kita-Kyushu lung cancer antigen-1.

DISCUSSION

KK-LC-1 is a cancer/testis antigen that is not expressed in normal tissues except for the testis, but has been found in cancers of multiple organs[18,19]. It has been recognized as a tumor-targeting peptide with high clinical potential for the diagnosis, imaging, and bioavailability of GC owing to its cancer/testis antigen characteristics[10]. KK-LC-1 expression may be characteristic of early-stage GC cases with a favorable prognosis, as it has been associated with longer overall survival in patients exhibiting KK-LC-1 expression[20]. In a previous study, we identified a correlation between KK-LC-1 expression and H. pylori infection in GC. However, we observed higher KK-LC-1 expression in cases with successful H. pylori eradication, suggesting that KK-LC-1 expression may be regulated by pathways independent of H. pylori infection[14]. Therefore, we hypothesize that gastric mucosa, irrespective of H. pylori infection status, expresses KK-LC-1 when it reaches a carcinogenic or precancerous state and subsequently develops into GC[13,14].

Most GCs are caused by H. pylori infection[21], which damages the gastric mucosa, leading to gastritis, chronic atrophic gastritis, intestinal metaplasia, and dysplasia, thereby increasing the risk of GC[21,22]. Some reports have suggested that H. pylori eradication reduces the incidence of metachronous carcinoma after endoscopic resection[2], whereas others report no preventive effect, especially when the gastric mucosa has progressed to the intestinal metaplasia or dysplasia stage[21,22], rendering the issue controversial[22]. This study serves as a preliminary investigation into the potential role of KK-LC-1 expression as a biomarker for the development of multiple GCs, providing a foundation for future comprehensive studies.

The expression rate of KK-LC-1 in multiple GC was 79.1%, consistent with previous reports. Among patients with multiple cancers, KK-LC-1 expression was positive in at least one lesion in 94.9% of cases. Although prior studies hypothesized that H. pylori infection induces KK-LC-1 expression[13], we observed KK-LC-1 expression in H. pylori-uninfected, currently infected, and previously infected lesions. Therefore, KK-LC-1 expression may occur during carcinogenesis independently of H. pylori infection[14]. Notably, in this study, the two cases lacking KK-LC-1 expression in two lesions were both H. pylori-uninfected, with initial lesions located in the upper region of the stomach. While there may be some relationship between H. pylori infection and KK-LC-1 expression, the small sample size limits definitive conclusions, warranting further research.

When examining KK-LC-1 expression in initial lesions alongside clinicopathological factors, differentiation type, multiple cancers, and H. pylori eradication were significantly associated with higher KK-LC-1 expression, consistent with previous findings[14]. To evaluate KK-LC-1 as a biomarker to identify the risk of multiple cancers, we analyzed characteristics of initial lesions. In cases where the initial lesion was located in the M&L stomach regions, KK-LC-1 expression was significantly higher among older patients, those with differentiated histology, and those with multiple cancers. KK-LC-1-positive tumors were detected in the initial lesions of all patients (100%) with multiple GCs in the middle and lower region, with a significantly higher incidence compared to that of patients with solitary cancers. Conversely, none of the patients with KK-LC-1-negative initial lesions in the middle and lower region developed multiple GCs, suggesting KK-LC-1 negativity may indicate a very low risk of metachronous or synchronous GC, although this finding is limited by sample size. In contrast, no similar association was observed for initial lesions located in the upper region.

This study has several limitations. It is a single-center study with a relatively small sample size and short observation period (3 years and 2 months), limiting the ability to capture all cases of multiple GCs. Future long-term follow-up studies with larger cohorts and detailed analyses of KK-LC-1 expression may clarify whether KK-LC-1 serves as a reliable risk factor for the development of multiple GCs.

CONCLUSION

Investigating KK-LC-1 expression in GC lesions may be useful in predicting multiple GCs. Specifically, if KK-LC-1 expression is positive in the middle and lower region, it may indicate the potential for the future development of multiple GCs and could function as a biomarker for predicting multiple GCs. Notably, in this study, none of the patients with KK-LC-1-negative initial lesions located in the middle and lower region developed multiple GCs. This finding suggests that the absence of KK-LC-1 expression in middle and lower lesions may be associated with an extremely low, potentially negligible, risk of developing multiple GCs.

ACKNOWLEDGEMENTS

The authors thank Yoshie Muraishi for their technical assistance.