PD-1, PD-L1, and CD163 in pancreatic undifferentiated carcinoma with osteoclast-like giant cells: expression patterns and clinical implications

Abstract

Undifferentiated carcinoma with osteoclast-like giant cells (UCOGC), a variant of pancreatic ductal adenocarcinoma (PDAC), has a striking genetic similarity to PDAC but a significantly improved overall survival. We hypothesize that this difference could be due to the immune response to the tumor, and as such, we investigated the expression of PD-1, PD-L1, and CD163 in a series of UCOGC. To this aim, 27 pancreatic UCOGCs (11 pure and 16 PDAC-associated), 5 extrapancreatic tumors with osteoclast-like giant cells and 10 pancreatic anaplastic carcinomas were immunostained using antibodies against PD-1, PD-L1, and CD163. In pancreatic UCOGCs, PD-L1 was expressed in neoplastic cells of 17 (63%) of 27 cases, more often in cases with an associated PDAC (P = .04). Expression of PD-L1 was associated with poor prognosis, confirmed by multivariate analysis: patients with PD-L1-positive UCOGCs had a risk of all-cause mortality that was 3 times higher than did patients with PD-L1-negative UCOGCs (hazard ratio, 3.397; 95% confidence interval, 1.023-18.375; P = .034). PD-L1 expression on tumor cells was also associated with aberrant P53 expression (P = .035). PD-1 was expressed on rare lymphocytes in 12 UCOGCs (44.4%), mainly located at the tumor periphery. CD163 was expressed on histiocytes, with a diffuse and strong staining pattern in all UCOGCs. Extrapancreatic tumors with osteoclast-like giant cells showed very similar staining patterns for the same proteins. Anaplastic carcinomas have some similarities to UCOGCs, but PD-L1 has no prognostic roles. Our results may have important implications for immunotherapeutic strategies in UCOGCs; these tumors may also represent a model for future therapeutic approaches against PDAC.

ESMO recommendations on microsatellite instability testing for immunotherapy in cancer, and its relationship with PD-1/PD-L1 expression and tumour mutational burden: a systematic review-based approach

“…The literature search identified globally 158 unduplicated articles, 126 of which were excluded after reviewing their title/abstract, which left 32 articles eligible for full text review. After applying our inclusion and exclusion criteria, 18 articles (for 17 cohorts) were selected for this systematic review [7,[59][60][61][62][63][64][65][66][67][68][69][70][71][72][73][74][75] (supplementary Figure S1, available at Annals of Oncology online).…”

“…The vast majority of the cohorts reported data on patients from the USA, followed by Asian and European subjects. The majority of the manuscripts (59%) investigated original cohorts of patients [7,61,[63][64][65][66][67][68][69][70]75], whereas data derived from external (public or private) databases were used in some papers [59,62,[71][72][73][74].…”

“…Complete data about TMB, MSI and PD-1/PDL1 of one cohort of patients was reconstructed by merging two studies on the same tumours [68,69].…”

“…The most used clone for immunostaining of PD-1 was NAT105 (60% of studies), and for PD-L1 was SP142 (47% of studies). The methods to assess the positivity for such markers differed among the selected studies, the most commonly applied being the simple reporting of the percentage of positive cells for PD-1 [60,61,68] and the use of an integrated threshold, coupling the intensity of staining and the percentage of positive neoplastic cells, for establishing PD-L1 positivity [7, 60-62, 64, 67, 68, 72-75]. To avoid reporting data affected by any bias, the quality of the studies was also evaluated using Newcastle-Ottawa scale (NOS) [76].…”

Pancreatic undifferentiated carcinoma with osteoclast-like giant cells curatively resected after pembrolizumab therapy for lung metastases: a case report

“…Different antibodies and sets of assay condition for PD-L1 have been used for respective immune checkpoint inhibitors with contrasting results [20,21], while the most used clone for PD-L1 is SP142 [11]. Luchini et al used clone E1L3N for PD-L1 IHC in above mentioned study [19], whereas we used 22c3 which is approved to aid in selection of patients for pembrolizumab treatment based on PD-L1 expression in FFPE tumor samples [22]. It is reported that 22c3 was prone to show lower mean score of PD-L1 expression in IHC for NSCLC compared to E1L3N [21], therefore it is unlikely that our results of PD-L1 high expression in IHC was false positive.…”