Prognostic Value of Ki67 Percentage, WT-1 Expression and p16/CDKN2A Deletion in Diffuse Malignant Peritoneal Mesothelioma: A Single-Centre Cohort Study

Abstract

Diffuse malignant peritoneal mesothelioma (DMPM) is a rare malignant neoplasm with a poor survival. Although some advances in knowledge have been obtained for the pleural form, much less is known about DMPM. Advantages in terms of prognosis are still limited and strong efforts need to be made. The aim of our study was to correlate several histological and molecular factors with survival in a large cohort of 45 DMPMs. We evaluated histotype, nuclear grade, mitotic count, necrosis, inflammation, desmoplastic reaction, Ki67 percentage, WT-1 expression, p16 protein by immunohistochemistry and CDKN2A deletion by FISH. Our results showed that epithelioid histotype, nuclear grade 2, mitotic count ≤5 x mm², absence of desmoplasia and p16/CDKN2A deletion, low Ki67 value, and high WT-1 expression were correlated with the most prolonged survival (p = 0.0001). Moreover, p16 loss in immunohistochemistry reflected CDKN2A deletion detected with FISH, and both were correlated with the worst survival (p = 0.0001). At multivariate analysis, Ki67 value, WT-1 expression and p16/CDKN2A deletion emerged as independent prognostic factors (p = 0.01, p = 0.0001 and p = 0.01, respectively). These parameters are easy to analyse at the time of DMPM diagnosis and may support better patient stratification, prediction of treatment effectiveness and therapeutic optimization.

Keywords: CDKN2A; Ki67; WT-1; diffuse malignant peritoneal mesothelioma; p16; prognostic factors.

Figure 1

Correlation between histological variables and survival. Sarcomatoid histotype (hematoxylin and eosin stain, original magnification ×200). (a), nuclear grade 3 (hematoxylin and eosin stain, original magnification ×200); (b), mitotic count >5 x mm² (hematoxylin and eosin stain, original magnification ×200); (c) and prominent desmoplastic stromal reaction (hematoxylin and eosin stain, original magnification ×400); (d) showed a shorter survival when compared with their counterparts (Green lines for: biphasic histotype in (a), nuclear grade 3 in (b), mitotic count >5 x mm² in (c), absence of desmoplasia in (d). Blue lines for: epithelioid histotype in (a), nuclear grade 2 in (b), mitotic count ≤5 x mm² in (c), presence of desmoplasia in (d). Yellow line for sarcomatoid histotype in (a)).

Figure 1

Correlation between histological variables and survival. Sarcomatoid histotype (hematoxylin and eosin stain, original magnification ×200). (a), nuclear grade 3 (hematoxylin and eosin stain, original magnification ×200); (b), mitotic count >5 x mm² (hematoxylin and eosin stain, original magnification ×200); (c) and prominent desmoplastic stromal reaction (hematoxylin and eosin stain, original magnification ×400); (d) showed a shorter survival when compared with their counterparts (Green lines for: biphasic histotype in (a), nuclear grade 3 in (b), mitotic count >5 x mm² in (c), absence of desmoplasia in (d). Blue lines for: epithelioid histotype in (a), nuclear grade 2 in (b), mitotic count ≤5 x mm² in (c), presence of desmoplasia in (d). Yellow line for sarcomatoid histotype in (a)).

Figure 1

Correlation between histological variables and survival. Sarcomatoid histotype (hematoxylin and eosin stain, original magnification ×200). (a), nuclear grade 3 (hematoxylin and eosin stain, original magnification ×200); (b), mitotic count >5 x mm² (hematoxylin and eosin stain, original magnification ×200); (c) and prominent desmoplastic stromal reaction (hematoxylin and eosin stain, original magnification ×400); (d) showed a shorter survival when compared with their counterparts (Green lines for: biphasic histotype in (a), nuclear grade 3 in (b), mitotic count >5 x mm² in (c), absence of desmoplasia in (d). Blue lines for: epithelioid histotype in (a), nuclear grade 2 in (b), mitotic count ≤5 x mm² in (c), presence of desmoplasia in (d). Yellow line for sarcomatoid histotype in (a)).

Figure 2

Correlation between immunohistochemical and molecular variables and survival. Ki67 > 25% (immunohistochemistry, original magnification ×400). (a) and WT-1 ≤ 15% (immunohistochemistry, original magnification ×400); (b) showed a shorter survival, whereas high p16 immunostaining score (immunohistochemistry, original magnification ×100); (c) and/or absence or heterozygous CDKN2A deletion showed a longer survival when compared with their counterparts; (d) (Green lines for: Ki67 > 25% in (a), WT-1 > 25% in (b), score 0–1 of p16 in (c), homozygous deletion of CDKN2A in (d). Blue lines for: Ki67 ≤25% (a), WT-1 ≤25% in (b), score 2–3 of p16 in (c), no/heterozygous deletion of CDKN2A in (d)).

Figure 2

Correlation between immunohistochemical and molecular variables and survival. Ki67 > 25% (immunohistochemistry, original magnification ×400). (a) and WT-1 ≤ 15% (immunohistochemistry, original magnification ×400); (b) showed a shorter survival, whereas high p16 immunostaining score (immunohistochemistry, original magnification ×100); (c) and/or absence or heterozygous CDKN2A deletion showed a longer survival when compared with their counterparts; (d) (Green lines for: Ki67 > 25% in (a), WT-1 > 25% in (b), score 0–1 of p16 in (c), homozygous deletion of CDKN2A in (d). Blue lines for: Ki67 ≤25% (a), WT-1 ≤25% in (b), score 2–3 of p16 in (c), no/heterozygous deletion of CDKN2A in (d)).