Aberrant expression of N-glycolyl GM3 ganglioside is associated with the aggressive biological behavior of human sarcomas

A number of 50 formalin-fixed and paraffin-embedded specimens from patients diagnosed with sarcoma, who underwent tumor surgical resection at the “Hermanos Ameijeiras” General Hospital (Havana, Cuba) between 2006 and 2013, were included. Samples from adult patients of any age, sex, race, histological subtype (except GIST) and any stage of disease were included in the study. Samples from patients who were treated or followed up in other medical institution were excluded.

All cases were staged according to the TNM classification established by the American Joint Committee on Cancer (AJCC) [12]. Performance status was evaluated using the ECOG score [13]. Clinical data such as age, gender, tumor size and localization, depth of tumor, the presence of metastasis, disease stage, recurrence, histological subtype and grade of differentiation were obtained from patient records. Overall survival (OS) was measured from the date of surgery to death for any cause or last follow-up and were as calculated for all patients.

This research was conducted after receiving the approved consent by the institutional Ethical Committee.

Sarcomas were immunohistochemically evaluated for a panel including but not limited to vimentin, desmin, pan-actin, S-100 protein, epithelial membrane antigen (EMA), cytokeratins, HNK-1 (CD57), protein gene product (PGP 9.5), CD99, CD34, c-kit, platelet-derived growth factor receptor (PDGFR), CD68 (MIC2), myogenic regulatory protein (MyoD1), h-caldesmon (HCD), alpha-1 antitrypsin (A1AT), smooth muscle actin (SMA), muscle specific actin (MSA), myoglobin, myogenin, and Ki-67 (MIB-1). For N-glycolyl GM3 ganglioside, the 14F7 Mab (a highly specific IgG1 against this molecule) produced at the Center of Molecular Immunology (Havana, Cuba) was used [11].

The method previously described [7], was used. Briefly, five-micrometer serial sections from each block were obtained, and the slides were dewaxed in xylene and rehydrated in ethanol following standard procedures. Afterward, the samples were incubated with 14F7 Mab followed by a peroxidase avidin-biotin system. Negative controls were performed substituting primary antibody for washing buffer (TBS). As positive controls, sections of breast adenocarcinoma with known positivity for NeuGcGM3 were used. Enzymatic activity was visualized with a DAB solution and slides were counterstained with Mayer’s Hematoxylin.

Immunohistochemical results were analyzed for both proportion of stained cells and intensity of 14F7 Mab reactivity. The percentage of positive cells was graded on a scale of 0–3 (0, no staining; 1, 1–50%; 2, 51–75%; and 3, 76–100%). The intensity of reaction was graded on a scale of 0–3; 0, no staining; 1, weak staining; 2, moderate staining; and 3, strong staining. Afterward, an immunoreactive scoring (IRS) was obtained by multiplying the two previously mentioned parameters. Finally, the expression of NeuGcGM3 was divided into low level (IRS < 6) or high level (IRS ≥ 6). For Ki-67 antigen, five areas of greater intensity of staining were subjectively selected and positive and negative cells was counted. Subsequently, the percentage of positive cells (positively stained nuclei/total number of cells × 100) was determined and the measure was grouped as follows: low cell proliferation rate (≤ 10%), moderate cell proliferation rate (11–50%), and high cell proliferation rate (≥ 50%). All slides were assessed by two trained observers (LS, NC) who did not have knowledge of clinical characteristics or outcomes.

Table 1 shows a summary of clinicopathological features of patients. The median of patient age at presentation was 51 years (ranged from 18 to 82 years). At the time of diagnosis, 19 (38.0%) and 31 (62.0%) of patients presented with 0 and ≥ 1 ECOG performance status, respectively. Seven of the 50 patients (14.0%) showed the presence of metastasis.

Median follow-up for the entire cohort was 2.21 years (range 0.17 to 9.50). After 3-years of follow-up, 22/50 (44.0%) of patients displayed tumor recurrence and 21/50 (42.0%) died within 5 years after diagnosis. According to the first-line therapy used, 37 (74.0%), 9 (18.0%) and 4 (8.0%) of cases received surgery, chemotherapy and supportive care, respectively. The scheme of chemotherapy used as a neoadjuvant and adjuvant regimen were: Ifosfamide/Doxorubicin, Doxorubicin, Scheme P6, Trabectedin, Gemcitabine/Docetaxel and Ifosfamide/Paclitaxel. Adjuvant radiation therapy was delivered in 19 (38%) patients.

The expression of NeuGcGM3 was observed in all cases, although a variable intensity and percentage of positive cells was evidenced (Fig. 1a and b). No alterations in the expression pattern of NeuGcGM3 was evidenced when primary treatment options were compared, as previously described by Blanco et al.. The staining was detected on both the membrane and cytoplasm of tumor cells with a finely granular staining pattern. The majority of samples had strong intensity (54.0%) and more than 50% of positive cells (92.0%) as shown in Table 2. A strong correlation was found between the percentage of positive tumor cells and staining intensity (Spearman’s correlation coefficient 0.564; p < 0.0001). According to the immunoreactive scoring (IRS) 33/50 (66.0%) of cases showed high levels of NeuGcGM3 expression.

The results of univariate and multivariate survival analysis are summarized in Table 3. In survival analysis, there was a statistically significant difference in the 5-year OS rates between high and low expression of NeuGcGM3 (45.4% vs. 82.3%; p = 0.016), while no significant relation was obtained with Disease-free survival (p = 0.346). In addition, univariate analysis showed that TNM stage (p = 0.000), occurrence of metastasis (p = 0.000), and expression of NeuGcGM3 (p = 0.034) (Fig. 1c) were significant prognostic factors for OS, while a tendency for association was evidenced for a histological grade of tumors (p = 0.091). Among these variables, only the presence of metastasis (HR = 11.34; 95% CI 2.60–49.4; p = 0.001) was an independent prognostic factor on multivariate analysis.

The relation of NeuGcGM3 expression with clinicopathological characteristics of patients is shown in Table 4. The level of immunoreactivity (IRS) was associated with age (p = 0.014). Interestingly, the presence of this ganglioside was statistically significant increased in advanced clinical stages (p = 0.022), high histological grade tumors (p = 0.013) as well as in samples displaying an increased index of cell proliferation (p = 0.012) (Figs. 2a, b and c). In addition, the expression of NeuGcGM3 was also increased in deeply located tumors when compared with superficial malignancies. However, only a tendency for association was obtained (p = 0.070). No significant differences were observed with the rest of the clinicopathological parameters.