Dominick Leone

Microvessel density, lymphovascular density, and lymphovascular invasion in primary cutaneous melanoma - Correlation with histopathologic prognosticators and BRAF status

The relationship between microvessel density (MVD), lymphovascular density (LVD), and lymphovascular invasion (LVI) in primary cutaneous melanoma (PCM) remains unclear. Given this, a total of 102 PCMs were assessed for MVD (vascular endothelial growth factor receptor 2 and Endocan), LVD (D2-40), and LVI (immunostaining with D2-40/S-100 and hematoxylin and eosin); tumoral S-100A13, vascular endothelial growth factor receptor 2, and Endocan; and BRAF status. LVD was associated with MVD (P = .01). MVD was higher in PCMs with depth greater than or equal to 2 mm and ulceration (P = .04, .05), whereas LVD was higher in PCMs with depth greater than or equal to 2 mm and mitoses (P = .03, .02). After adjusting for MVD and LVD, only ulceration was associated with LVI (P < .02). A BRAF mutation was seen in 30.4% cases, and when present, both LVD and host response (P = .0008 and .04, respectively) were significantly associated with MVD. Immunostaining with S-100A13 was noted in 99% of cases and a significant association noted only with ulceration (P = .05). Immunostaining increased LVI positivity (46.5% versus 4.9% by hematoxylin and eosin, P < .0001). MVD and LVD are not associated with LVI, appear to be closely related with each other, and are associated with select markers of poor prognosticative value. The association between a host response and LVD and MVD in PCMs with a BRAF mutation suggests that they exhibit potential for strategizing immunotherapies.

Tumoral PD-L1 expression in desmoplastic melanoma is associated with depth of invasion, tumor-infiltrating CD8 cytotoxic lymphocytes and the mixed cytomorphological variant

Recently, patients with metastatic desmoplastic melanoma (DM) have been shown to respond more favorably to anti-PD1/PD-L1 therapy than other melanoma subtypes. Given this, we evaluated PD-L1/2 expression in primary DM samples and correlated these with subtype, CD8+ lymphocyte status, histopathological prognosticators, and select genetic alterations. Eighty-six (36 mixed DM, 50 pure DM) archival annotated samples met inclusion criteria and were immunohistochemically semiquantitatively evaluated. Per established criteria, for PD-L1/L2, cases with ⩾5% tumoral expression, and for CD8, cases with a predominantly peri/intratumoral CD8+ infiltrate were scored positive. Univariate analysis (chi-square and Wilcoxon) identified potential confounders and a nested case–control study was accomplished using multiple logistic regression. For PD-L1, 49% of cases were positive and 71% of cases with thickness >4 mm were positive; PD-L1 expression differed by median depth (3.29 mm, interquartile range=3.58 mm for PD-L1 positives vs 1.75 mm, interquartile range=2.04 mm for PD-L1 negatives, P=0.0002) and was linearly associated with increasing depth of invasion (P=0.0003). PD-L1-positive cases were more likely to display CD8+ lymphocytes (60 vs 28% P=0.0047).The presence of CD8+ lymphocytes correlated significantly with depth of invasion >1 mm (P=0.022). On multivariate analysis, PD-L1 was 6.14 × more likely to be expressed in mixed DM than pure DM (P=0.0131), CD8+ staining was 6.22 × more likely in PD-L1 positive cases than in PD-L1 negative (P=0.0118), and tumor depth was associated with greater odds of PD-L1 expression (OR=1.61, P=0.0181). PD-L2 expression was observed in 48% of cases but did not correlate with any variables. Correlation of tumoral PD-L1 with increased depth and CD8+ lymphocytes implicates the tumoral immune microenvironment with advancing disease in DM. Enhanced tumoral PD-L1 expression in the mixed cytomorphological variant provides an insight into the differential pathogenesis of the subtypes and suggests that these patients are likely better candidates for anti-PD/PD-L1 therapy.

Neurotrophin receptors and perineural invasion in desmoplastic melanoma

BACKGROUND Perineural invasion (PNI) in desmoplastic melanoma is associated with increased local recurrence and reduced disease-free survival. The biological mechanisms underlying PNI remain unclear although several lines of evidence implicate neurotrophins and their receptors. OBJECTIVES We investigated the expression of p75NGFR and TrkA, and the presence of functional RET polymorphism (RETp) as they relate to PNI in desmoplastic melanoma. METHODS In all, 43 cases of desmoplastic melanoma were immunohistochemically evaluated for TrkA and p75NGFR expression and RETp was detected by direct DNA sequencing. RESULTS PNI was present in 67% of cases. On univariate analysis, p75NGFR was associated with PNI (expression detected in 79% of PNI-positive cases compared with 36% of PNI-negative cases, P = .005), increased Breslow depth (P = .007), and greater Clark level (P = .01). RETp was noted in 28% of cases but was not significantly associated with PNI (P = .27) or other histopathologic variables. TrkA expression was absent in all cases. PNI was associated with increased Breslow depth and Clark level (P = .01 and P = .009, respectively). Controlling for the association between p75NGFR and depth, p75NGFR remained associated with an increased propensity for PNI (odds ratio 4.68, P = .04). LIMITATIONS The sample size was limited. CONCLUSION In desmoplastic melanoma, p75NGFR expression is significantly associated with PNI and a more locally aggressive phenotype.

Protein expression of the chemokine receptor CXCR4 and its ligand CXCL12 in primary cutaneous melanoma—biomarkers of potential utility? ☆ ☆☆

Dysregulation of the CXCR4/CXCL12 axis, relevant in melanoma progression, activates cell cycle progression and migration via stimulation of the MAPK pathway. We sought to ascertain the cooperativity of the CXCR4/CXCL12 axis with established prognosticators and BRAF status in melanoma. Samples (n = 107) of primary cutaneous melanoma were assessed for protein expression of CXCR4 and CXCL12, and molecular analyses were performed to ascertain BRAF status. Univariate analyses of CXCR4 protein showed that the proportion of CXCR4 positives was greater in melanomas with absence of mitoses (P < .0001), absence of ulceration (P = .0008), and absence of regression (P = .02). Patients presenting at shallower stages (American Joint Committee on Cancer [AJCC] 1-2) exhibited a larger proportion of CXCR4 positives (76.9%, P < .0001 and 69.0%, P = .008), whereas those at deeper stages (AJCC 3-4) exhibited a larger proportion of negatives (75.0%, P = .004 and 66.7%, P = .22). In a multivariate analysis, lower odds of CXCR4 protein expression were associated with AJCC stage 3 (odds ratio [OR]=0.16, P = .01), AJCC stage 4 (OR=0.17, P = .04), and mitoses (OR=0.21, P = .01). Univariate analyses of CXCL12 protein showed that the proportion of CXCL12 negatives was significantly smaller in melanomas with depth of at least 1 mm, absence of ulceration, and absence of vascular invasion (P < .0001 for all). CXCR4 and CXCL12 appear to be biomarkers associated with established prognosticators of good and poor clinical outcome, respectively, in primary cutaneous melanoma. A BRAF mutation does not appear to be associated with CXCR4/CXCL12 axis upregulation in primary cutaneous melanoma.

Neurofibromin protein loss in desmoplastic melanoma subtypes: implicating NF1 allelic loss as a distinct genetic driver?

Loss of the NF1 allele, coding for the protein neurofibromin, and polymorphism in the proto-oncogene RET (RETp) are purportedly common in desmoplastic melanoma (DM). DM is categorized into pure (PDM) and mixed (MDM) subtypes, which differ in prognosis. Most NF1 mutations result in a truncated/absent protein, making immunohistochemical screening for neurofibromin an ideal surrogate for NF1 allelic loss. Using antineurofibromin, our aims were to ascertain the incidence of neurofibromin loss in DM subtypes and to evaluate the relationship with RET, perineural invasion (PNI) and established histopathologic prognosticators. A total of 78 archival samples of DM met criteria for inclusion (54 cases of non-DM serving as controls). Immunohistochemistry was performed for neurofibromin, whereas direct DNA sequencing was used for RETp and BRAF mutation status. Statistical analyses included χ(2) test as well as Fisher exact test. Neurofibromin loss was more common in DM than non-DM (69% versus 54%; P=.02). In DM, significant differences in neurofibromin loss were noted in the following: non-head and neck versus head and neck biopsy site (88% versus 55%) and PDM versus MDM variants (80% versus 56%). No significant associations were noted with sex, presence of a junctional component, Breslow depth, ulceration, mitoses, host response, RETp, BRAF status, or PNI. RETp was marginally associated with PNI-positive DM versus PNI-negative DM (36 versus 18%; P=.08). Our findings, the largest to date investigating neurofibromin in DM, validate the incidence of NF1 mutations/allelic loss in DM and suggest that the DM subtypes have distinct genetic drivers.

Perineural invasion in cutaneous squamous cell carcinoma: role of immunohistochemistry, anatomical site, and the high-affinity nerve growth factor receptor TrkA

Perineural invasion (PNI) has been recently added to the American Joint Committee on Cancer cutaneous squamous cell carcinoma (cSCC) staging criteria as a high-risk tumor characteristic and is purportedly more common in cSCCs of the head and neck (H&N). Expression of the high-affinity nerve growth factor receptor TrkA has been shown to be associated with PNI in noncutaneous neoplasms. Given this, we sought to ascertain the incidence of PNI in cSCCs using double immunostaining (DIS) and to investigate PNIs relationship with TrkA and established histopathologic prognosticators. Fifty-seven cSCCs from the H&N and 53 from non-H&N areas were immunohistochemically analyzed for PNI (DIS with S-100 and p63) and TrkA expression. Comparing H&N versus non-H&N areas, using hematoxylin and eosin, PNI was detected in 11% versus 6% cases, respectively, and, using DIS, in 23% versus 15%, respectively, with significant disagreement between both methods (κ = 0.47; P = .002). There was a 2.33-fold increase in PNI detection with DIS compared to hematoxylin and eosin (95% confidence interval, 1.12-4.87; P = .02). TrkA expression was 1.96 times more frequently observed in cSCCs from the H&N compared to those from non-H&N areas (P = .01). Regardless of site, TrkA expression was associated with decreased degree of differentiation (odds ratio, 6.46; P = .0006) and high-risk morphologic variants (odds ratio, 6.53; P = .002) but not significantly associated with PNI (P = .33). Increased PNI detection with DIS underscores the adjunctive utility of immunohistochemistry in microstaging. Significantly more common TrkA expression in cSCCs of the H&N argues in favor of heterogeneity among SCCs from different anatomical sites.

Frequency of telomerase reverse transcripter promoter mutations in desmoplastic melanoma subtypes: analyses of 76 cases.

Estimates of the frequency of telomerase reverse transcripter (TERT) mutations in desmoplastic melanoma (DM) are limited. DM is categorized into subtypes, pure and mixed, differing in prognosis, suggesting genetic heterogeneity. Given this, our aims were to determine the incidence of TERT promoter mutations in DM subtypes and to evaluate its relationship with established histopathologic prognosticators, BRAF and RETp status, and neurofibromin protein expression. Of the archival annotated samples retrieved, 76 cases of DM (48 pure and 28 mixed) fulfilled the criteria for inclusion. PCR amplification of the TERT promoter region was performed on DNA extracted from formalin-fixed paraffin-embedded tissue using primers5′-GCCGATTCGACCTCTCTCC-3′ (forward) and 5′-CAGCGCTGCCTGAAACTC-3′ (reverse). For each case, appropriate C>T mutations were identified on the electropherograms. Univariate analysis using &khgr;2-test was carried out to identify potential confounders; a nested case–control study of demographic, clinical, histopathological, and genetic determinants was carried out using multiple logistic regression. Significant differences in TERT promoter mutation frequencies were noted in the subtypes (mixed vs. pure; 15/28, 54% vs. 11/48, 23%, respectively, P=0.0066). After adjusting for potential confounding, multivariate analyses indicated a three-fold increase in the odds of the TERT mutation for those with the mixed subtype compared with the pure subtype (P=0.04, adjusted odds ratio =3.32). No other significant associations were noted (sex/junctional component/Breslow depth/ulceration/mitoses/host response/RETp, BRAF status, and neurofibromin protein expression). Our findings, the largest to date investigating TERT promoter mutations in DM, support the hypothesis that the subtypes have distinct genetic drivers and underscore the relevance of telomere integrity in the etiopathogenesis of the mixed variant.

Correlation of chemokine receptor CXCR4 mRNA in primary cutaneous melanoma with established histopathologic prognosticators and the BRAF status.

Dysregulation of the chemokine receptor CXCR4 is relevant in melanoma progression, and the CXCR4/CXCL12 axis has been shown to activate cell cycle progression and malignant cell migration through stimulation of the mitogen-activated protein kinase pathway. Studies ascertaining the potential utility of CXCR4 mRNA as a prognosticator in melanoma have focused mainly on metastatic melanoma with conflicting results. In the light of this, we sought to explore the potential relationship between CXCR4 mRNA expression with established histopathologic prognosticators and BRAF status in melanoma. Archived consecutive samples (n=107) of primary cutaneous melanoma were retrieved and assessed for the following: CXCR4 mRNA (semiquantitative RT-PCR) and BRAF exon 15 status (DNA Sanger sequencing). Statistical analyses included correlation between CXCR4 mRNA levels and established histopathologic prognosticators as well as the BRAF status using univariate and multiple linear methods. Multivariable analyses revealed a significant correlation between elevated CXCR4 mRNA (low &Dgr;Ct value) and the presence of BRAF mutation (P=0.02). Absence of a brisk host response was associated with elevated CXCR4 mRNA expression (P=0.04). CXCR4 mRNA was significantly lower in AJCC stage 2 compared with stage 1 after controlling for significant clinical prognosticators (P=0.02). The association between elevated CXCR4 mRNA and absence of a brisk host response suggests that CXCR4 may be involved in regulation of the host immune response in melanoma and is a molecule of potential utility as a biomarker for recruiting melanoma patients for immunotherapy. Higher CXCR4 mRNA in patients with a BRAF mutation suggests its utility as a putative therapeutic target.

BRAF and epithelial-mesenchymal transition in primary cutaneous melanoma: a role for Snail and E-cadherin?

In vitro studies in melanoma indicate that up-regulation of the transcriptional repressor Snail occurs with a concomitant decrease of its target E-cadherin, both hallmarks of epithelial-mesenchymal transition-an association not established in vivo. We sought to elucidate the relationship between BRAF, Snail, E-cadherin, and established histopathologic prognosticators in primary cutaneous melanoma. Archived annotated samples with a diagnosis of primary cutaneous melanoma were retrieved (n = 68 cases; 34 BRAF mutant and 34 BRAF wild type) and immunohistochemically stained for Snail and E-cadherin protein expression. A semiquantitative scoring system was used. Multivariate logistic analysis was used to control confounders of BRAF. Snail expression was significantly associated only with ulceration (42% versus 13%; P = .02). E-cadherin expression was present in 26% of BRAF mutant and 71% of BRAF wild-type cases (P = .0003). Loss of E-cadherin expression was associated with female sex (60% versus 34%; P = .05), BRAF mutation (74% versus 29%; P = .0003), thickness greater than or equal to 1 mm (68% versus 32%; P = .004), mitosis (63% versus 25%; P = .007), and ulceration (75% versus 44%; P = .05). BRAF mutation was associated with male sex (60% versus 30%; P = .02), Breslow thickness (P = .007), thickness greater than or equal to 1 mm (68% versus 29%; P = .002), and ulceration (75% versus 42%; P = .02). Snail expression did not correlate with loss of E-cadherin expression (47% versus 53%; P = .79). After controlling for potential confounding, BRAF mutation was associated with loss of E-cadherin (adjusted odds ratio, 8.332; 95% confidence interval, 2.257-30.757; P = .0015) and Breslow thickness greater than 1 mm (adjusted odds ratio, 7.360; 95% confidence interval, 1.534-35.318; P = .0126). Our findings, indicating that mutant BRAF represses E-cadherin expression, implicating a catalytic role for BRAF in epithelial-mesenchymal transition.

Concordance of somatic mutation profiles [BRAF, NRAS and TERT] and tumoral PD-L1 in matched primary cutaneous and metastatic melanoma samples

Despite the efficacy of BRAF-targeted and PD-L1-related immune therapies in tackling metastatic melanoma, a significant number of patients exhibit resistance. Given this, the objective of the current study was to ascertain concordance of somatic mutations in BRAF/NRAS/TERT and immunohistochemical PD-L1 and CD8 in matched primary cutaneous and metastatic melanoma. A total of 43 archival paired samples with sufficient material for genetic and immunohistochemical analyses met the criteria for inclusion in the study. Immunohistochemistry was performed for PD-L1 and CD8 and direct-DNA Sanger sequencing for BRAF/NRAS/TERT promoter mutational analyses. Agreement between paired samples was assessed using Cohen κ. Poor concordance among primary and corresponding metastases was noted in BRAF (9/42 cases discordant, κ = 0.49; 95% confidence interval [CI], 0.21-0.77; P = .0013), TERT promoter mutations (13/41 cases discordant, κ = 0.33; 95% CI, 0.04-0.62; P = .033), tumoral PD-L1 immunoexpression (9/43 cases discordant, κ = 0.39; 95% CI, 0.07-0.72; P = .0099), and immunoexpression of CD8+ T lymphocytes (12/43 cases discordant, κ = 0.44; 95% CI, 0.19-0.69; P = .002). Although NRAS1 and NRAS2 were highly concordant (42/43 and 39/43 cases, respectively), discordant NRAS2 mutational status was associated with a median time to metastasis of 90 versus 455 days for pairs with concordant status (P = .07). Although limited by sample size, our findings suggest that consideration be given to mutational analysis of metastatic tissue rather than the primary to guide BRAF-targeted therapy and question the roles of TERT promoter mutations and PD-L1 as predictive biomarkers in malignant melanoma.