Association of chromosomal aberrations in chromosomes 3 and 7, and P16 mutations with malignancy in salivary gland tumors

Clinical samples

Surgically excised salivary gland tumor and benign/normal salivary gland tissue specimens were collected from the affiliated hospital of Guangdong Medical University. Based on the 5^th edition of the WHO classification for salivary gland tumors, the samples were categorized into 38 malignant and 40 benign salivary gland tumors or normal salivary gland tissues. The malignant salivary gland tumor samples included 20 cases of mucoepidermoid carcinoma, 13 of adenoid cystic carcinoma, two of salivary duct carcinoma, two of myoepithelial carcinoma, and one of secretory carcinoma, which is not classified in the 5^th edition of the WHO classification. The benign salivary gland tumor samples comprised 19 cases of pleomorphic adenoma, 11 of Warthin’s tumor, one of basal cell adenoma, and nine of benign/normal salivary gland tissue. Tumors of non-salivary gland origin were excluded. Tissue sections were cut to a thickness of 5 μm. Peripheral blood cells were collected from healthy volunteers with no history or presence of cardiovascular, renal, hepatic, pulmonary, hematological, gastrointestinal, metabolic, endocrine, immunological, neurological, or psychiatric diseases. All procedures adhered to the Declaration of Helsinki and were approved by the Ethics Committee of Guangdong Medical University (number PJKT2024-196).

Construction of FISH probes

The CSP3 and CSP7 probes were designed based on the high-specificity centromeric regions of human chromosomes 3 and 7. Primer sequences for the probes are listed in Table S1. The P16 gene sequence was selected from Invitrogen’s RP11 BAC clone library (Table S2), with plasmids extracted using a commercial kit (Takara Biomedical Technology, Dalian, China). Primers were synthesized by Sangon Biotech Co. Ltd. (Shanghai, China), with human genomic and plasmid DNA as the templates for polymerase chain reaction (PCR) amplification. The PCR products were analyzed using 2% agarose gel electrophoresis. The gap-filling method was adopted for fluorescent labeling of the amplified products, with tetramethylrhodamine (TAMRA, Roche Diagnostics, Indianapolis, IN, USA) as the fluorophore. The reaction mixture was prepared, mixed, centrifuged on ice under strict light protection, and labeled at 25 °C for 2 h. Enzyme inactivation occurred at 80 °C for 10 min. The labeled products were precipitated with sodium acetate and purified by high-speed centrifugation.

FISH analysis

Slides were preheated at 65 °C for 5–30 min. Deparaffinization was performed at room temperature using an environmentally friendly agent (Guangzhou Anbiping Pharmaceutical Technology Co., Ltd., Yuexiu, China) for 10 min. Slides were immersed in room-temperature anhydrous ethanol to remove residual agents for 10 min. Rehydration was performed sequentially in anhydrous ethanol, 90% ethanol, 70% ethanol, and purified water, each for 3 min. Samples were boiled in purified water at 95–100 °C for 25 min, air-dried, and digested with 100–200 μL pepsin working solution at 37 ± 1 °C for 3–15 min. After washing in 2 × SSC (saline-sodium citrate) at room temperature for 3 min, slides were dehydrated sequentially in 70%, 90%, and 100% ethanol at room temperature for 2 min each, then air-dried. Approximately 4–8 μL hybridization solution was applied to the hybridization area, covered with a coverslip, and gently pressed to ensure uniform probe distribution and bubble removal. The coverslip was sealed with rubber cement, denatured at 85 ± 1 °C for 5 min, and hybridized at 37 ± 1 °C for 10–18 h. After gently removing the rubber cement and coverslip, slides were washed at 2 × SSC at 37 ± 1 °C for 10 min, followed by washing in 0.1% nonylphenol (NP)-40/2 × SSC at 37 ± 1 °C for 5 min. The slides were immersed in 70% ethanol at room temperature for 3 min, then stained with 5–7 μL of 4′,6-diamidino-2-phenylindole (DAPI) after air-drying. The slides were examined under a fluorescence microscope after sealing with a coverslip in the dark.

Statistical evaluation

Fluorescent signals of DAPI and probes were observed, with CSP7 and GSP P16 probes emitting red fluorescence, while CSP3 probes emitted green fluorescence. Fifty cells with distinct cell contours, uniform DAPI staining, clear probe signals, and no overlapping nuclei were selected. The nucleus’s green and red fluorescent signals were counted at 1,000x magnification, and the ratio of abnormal to normal cells was calculated. Each sample was evaluated by counting the same number of cells, and the percentage of abnormal signal points was recorded. A sample was considered positive if the rate of abnormal signal points exceeded the threshold (10% for chromosome 3/7 abnormalities and 20% for P16 gene abnormalities); otherwise, it was deemed negative. If the percentage equaled the threshold, additional cell observations were required. Sensitivity was defined as the concordance rate between the probe-diagnosed positive samples and pathological diagnosis positive samples (malignant salivary gland tumors). Conversely, specificity was defined as the concordance rate between probe-diagnosed negative samples and pathological diagnosis negative samples (normal/benign salivary gland tissue).

Validation of CSP3, CSP7, and GSP P16 probes

To assess the sensitivity and specificity of the designed probes, they were tested on human peripheral blood cultured cells. Fifty metaphase cells were analyzed to evaluate the fluorescence intensity, hybridization efficiency, and accuracy of hybridization sites. The gene loci corresponding to CSP3 and GSP P16 probes were labeled with green and red fluorescence, respectively. The fluorescent signals were clear, with no cross-hybridization, indicating a 100% (50/50) hybridization efficiency (Figs. 1A–1D).

Application of CSP3, CSP7, and GSP P16 probes in salivary gland tumors

FISH probes were applied to 34 salivary gland tumor samples with confirmed pathological diagnoses, including 22 malignant tumors (mucoepidermoid carcinoma and adenoid cystic carcinoma) and 12 benign tumors. The results were as follows: CSP3 showed 17 true positives, five false positives, 10 true negatives, and two false negatives; CSP7 revealed seven true positives, 15 false positives, 11 true negatives, and one false negative; GSP P16 demonstrated 17 true positives, five false positives, 10 true negatives, and two false negatives (Table S3). The FISH results were analyzed (Table 1), with representative images displayed (Figs. 2A–2B & 3A–3D). Moreover, the sensitivity and specificity of the CSP3, CSP7, and GSP P16 probes, and their combinations in distinguishing benign from malignant tumors, were calculated (Table 2). The results indicated that CSP3 probes exhibited high specificity (>80%) for salivary gland tissues, whereas CSP7 probes had relatively low sensitivity (31.8%). Moreover, the CSP3 and GSP P16 combination yielded the best performance, with 90.9% sensitivity and 83.3% specificity, outperforming individual probes and other combinations, thereby offering superior detection for malignant salivary gland tumors.

The combination of CSP3 and GSP P16 probes in detecting malignant salivary gland tumors

The sample size was expanded to 78 to validate the sensitivity and specificity of CSP3, GSP P16, and their combination (Table 3). Among 38 malignant and 40 benign cases, CSP3 exhibited 23 true positives, 15 false positives, 36 true negatives, and four false negatives. Meanwhile, GSP P16 showed 28 true positives, 10 false positives, 34 true negatives, and six false negatives (Table S4). A sample was considered positive if the CSP3 or GSP P16 probe exhibited a fluorescent signal. The combined probes’ efficacy was assessed in the 78 clinical samples, with the results compared to actual pathological diagnoses. The receiver operating characteristic (ROC) analysis demonstrated sensitivities of 60.5% and 73.7%, specificities of 90.0% and 85.0%, and areas under the curves (AUCs) of 0.716 and 0.831 for CSP3 and GSP P16, respectively, in detecting malignant salivary gland tumors. When using the combined CSP3 and GSP P16 probes (with either probe positive), sensitivity increased to 89.5%, specificity remained at 85.0%, and the ROC area was 0.862 (Fig. 4).

The results indicate that the P16 gene probe exhibits higher sensitivity, while the CSP3 probe shows superior specificity when used individually. Surprisingly, the combination of CSP3 and GSP P16 probes enhances sensitivity for malignant tumors, albeit with a slight reduction in specificity, thereby improving overall detection efficacy for malignant salivary gland tumors.

The clinical data of 78 patients (Table S5) were analyzed, among which 44.87% (35/78) had salivary gland tumors located in the parotid gland, followed by the submandibular gland at 17.95% (14/78) (Fig. S1A). In the group of 38 patients with malignant tumors, the parotid gland remained the most commonly affected, comprising 34.21% (13/38), followed by the respiratory tract and submandibular gland at 15.79% (6/38) (Fig. S1B). When further classifying tumor locations, the major salivary glands, consisting of the parotid, submandibular, and sublingual glands, were revealed to be more frequently involved, accounting for 66.67% of all salivary gland tumors and 55.26% of malignant salivary gland tumors (Figs. S1C–S1D). Moreover, no substantial difference was observed in the age distribution between patients with benign (49.38; 95% confidence interval (CI) [43.47–55.28]) and malignant salivary gland tumors (48.74; 95% CI [39.18–53.42]) (Fig. S1E).