Background:
Atypical fibroxanthoma (AFX) and pleomorphic dermal sarcoma (PDS) represent a spectrum of cutaneous mesenchymal neoplasms with shared clinical, genetic, and histopathologic features. Preferentially expressed antigen in melanoma (PRAME), a cancer-testis antigen increasingly used as a diagnostic and prognostic marker in melanocytic tumours and selected sarcomas, has not been well characterised in AFX and PDS¹.

Purpose:
The aim of this study is to evaluate the pattern of PRAME staining across the AFX/PDS spectrum.

Methods:
A retrospective search of the Laboratory Information Management System for cases reported as AFX and/or PDS with accompanying PRAME immunohistochemistry between January 2023 and January 2026 at The Christie Hospital NHS Trust, a tertiary referral centre, was conducted. Thirty-one cases were identified; 19 had in-house PRAME slides available for review. Nuclear staining extent (0-4) and intensity (0-3) were assessed and documented using a proforma, generating a combined score (0-7).

Results:
All 19 reviewed cases (3 AFX/PDS, 3 AFX, 13 PDS) demonstrated PRAME expression. Staining was predominantly weak and patchy, with a mean intensity score of 1, mean extent score of 2, and mean combined score of 3 (range 2-5). Of the 12 non-reviewed cases, reports described ‘weak/patchy/focal’ staining in 7, ‘negative’ staining in 4, and ‘diffuse’ staining in 1 case; subsequent melanoma-targeted NGS in the latter identified no pathogenic variants.

Conclusions:
Although used primarily in the diagnosis of melanoma, PRAME has been shown to be variably expressed in carcinomas and sarcomas. In this study, PRAME showed patchy and weak positivity in most cases of AFX/PDS. This is significant as AFX/PDS is a diagnosis of exclusion and tends to be negative for most lineage-specific markers. In the absence of other lineage-specific markers, a positive PRAME should not preclude the diagnosis of AFX/PDS, especially if the staining pattern is patchy and/or weak.

Background: Under RCPath criteria, an unequivocal diagnosis of invasive colorectal adenocarcinoma on biopsy requires definite submucosal invasion. Superficial specimens may show high-grade dysplasia or suspicious features without demonstrable invasion, yielding non-definitive results with important management implications, including ancillary studies.
Purpose: To evaluate whether initial colonoscopic colorectal biopsies meet diagnostic standards, identify factors associated with inability to demonstrate invasion, and assess the clinical impact of non-definitive findings.
Methods: A single-centre retrospective standards-based audit of surgically resected colorectal adenocarcinomas was performed. The initial diagnostic biopsy for each case was classified as definitive or non-definitive for invasion according to RCPath criteria. Clinical, colonoscopic, biopsy, and resection tumour pathological variables were analysed, with significant univariate factors entered into multivariate logistic regression restricted to treatment-naïve cases to avoid therapy effects.
Results: Of 127 patients, 44 (35%) had non-definitive results. Excluding cases histologically compatible with invasion that were managed as malignant following multidisciplinary review left 39/127 patients (31%) without definitive histologic confirmation. Repeat biopsies further reduced this to 31/127 (24%). Demographic, clinical and operator variables were not associated with non-definitive biopsies. Among treatment-naïve patients, exophytic tumour morphology and histologic obscuring features (ulceration/necrosis) independently predicted non-definitive biopsies, whereas fragment number was not significant. Management varied by site: most colon cancers had direct surgery (25/27, 93%), while rectal cancers had repeat biopsy (9/16), neoadjuvant therapy (3/16), or direct surgery (4/16).
Conclusions: Nearly one quarter (24%) of colorectal cancers lacked definitive histologic confirmation after diagnostic work-up despite invasive adenocarcinoma at resection. Biopsy diagnostic yield is multifactorial, driven by gross tumour morphology and surface ulceration/necrosis rather than operator or sampling size, with greater clinical impact in rectal cancer pathways. Our findings reinforce the recognised limitations of superficial endoscopic biopsies in demonstrating invasion despite strong endoscopic suspicion of malignancy, with implications for colorectal cancer management in the neoadjuvant era.

Background: Papillary thyroid carcinoma (PTC) carries an excellent prognosis, yet lymph node metastasis (LNM) significantly affects prognosis. Tumour border configuration is an established prognostic factor in several malignancies, but has not been studied in PTC.

Purpose: To describe tumour border configuration and stromal reaction in PTC and to assess its association with lymph node metastasis.

Methods: This retrospective cross-sectional study included PTC cases diagnosed from 2021 to 2023 at the National Hospital Galle and Faculty of Medicine, University of Ruhuna. Cases with confirmed PTC and documented lymph node assessment were included. Encapsulated tumours were excluded. Tumour borders were classified as pushing or infiltrative. Presence of the stromal reaction recorded. Associations with LNM and lymphovascular invasion (LVI) were analysed.

Results: A total of 88 (mean age of 43.6 years, with 79.5% less than 55 years) PTC cases were included, with 78 (88.6%) females. Most tumours (64.8%) measured ≤10 mm. Conventional PTC accounted for 65.9% of cases, followed by follicular variant (30.7%). Infiltrative borders were identified in 61.4% (n = 54), while 38.6% (n = 34) showed pushing borders. Stromal reaction was present in 48.9% (n = 43). LVI was evident in 27.3% (n=24). LNM was identified in 15.9% (n = 14). Tumour border configuration was not significantly associated with LNM (p = 0.149) or LVI (p=0.53). Stromal reaction demonstrated a significant positive association with LVI (p = 0.003).

Conclusion: While infiltrative borders were more common, tumour border configuration was not significantly associated with LNM. However, the presence of stromal reaction at the invasive front had a significant positive association with LVI, suggesting a role in tumour aggressiveness. This highlights the possibility that the tumour–stroma interaction plays a role in early tumour spread.

Background

Regional lymph node status after surgery is a key prognostic factor in rectal cancer. Lymph node size and number partly reflect host immunity, but it is unclear whether these measures are modified by neoadjuvant radiotherapy and related to the degree of tumour regression.

Purpose

To investigate whether neoadjuvant short-course radiotherapy (SCRT) alters lymph node size and number in rectal cancer and whether these changes are associated with the degree of tumour regression.

Methods

Digital H&E slides from a randomly selected subset of the MRC CR07 trial comparing SCRT with straight to surgery (STS) were analysed. Lymph node area and diameter were measured using Aperio ImageScope and nodes were categorised as involved or non-involved by tumour. Groups were compared using the Mann Whitney U test.

Results

84 cases from 20 UK centres were included in the study. A total of 922 lymph nodes (146 involved, 778 non-involved) from 36 SCRT and 48 STS cases were analysed. Involved nodes were significantly larger than non-involved nodes in both trial arms (p<0.0001). No difference was observed between trial arms for total node number (SCRT=9 vs. STS=10, p=0.648) or involved node area (SCRT=23.7mm2 vs. STS=22.0mm2, p=0.441). However, a reduction in non-involved node diameter was observed following SCRT (3.4mm vs. 3.8mm, p=0.042) with a corresponding reduction in area (7.5mm2 vs. 10.3mm2, p=0.0572). In the SCRT group, good responders demonstrated smaller involved nodes (20 mm2 vs. 33 mm2, p=0.055).

Conclusions

This study suggests that SCRT reduces the size of non-involved lymph nodes, which may reflect the effect of radiotherapy on host immune response. In good responders, the size of involved nodes reduced, likely reflecting regression in the nodal metastatic disease. Further studies are required to confirm these findings and their relationship to longer term outcomes.

Acknowledgements

This research was supported by the PathSoc Undergraduate Elective Bursary.

Background:
The morphology of uninvolved lymph nodes (LN-neg), particularly the numbers of secondary lymphoid follicles (SLFs), may reflect important host anti-tumour responses in locally advanced, operable colon cancer. Recent pilot data from the FOxTROT trial demonstrated a potential association between LN-neg architectural features and survival.

Purpose:
This study aimed to investigate LN-neg in a second independent cohort of the FOxTROT clinical trial.

Methods:
152 additional randomly selected FOxTROT cases were included in the current study; 43 underwent straight-to-surgery (STS) and 109 received 6 weeks of neoadjuvant chemotherapy before surgery (NAC). Digitalised H&E slides were annotated to determine LN-neg area and diameter. SLFs were counted in each node and SLF density calculated from the SLF count divided by the LN-neg area.

Results:
LN-neg number remained high and did not differ between arms (median STS=19 [IQR 15-23] vs. NAC=18 [13-23]; p=0.35). NAC was associated with a smaller LN-neg area (STS=14.7 mm² [10.5-18.3 mm2] vs. NAC=10.7 mm² [7.0-14.5 mm2]; p=0.0015), smaller LN-neg diameter (STS=4.7 mm [3.9-5.2 mm] vs. NAC=4.3 mm [3.3-5.0 mm]; p=0.041), and fewer LN-neg SLFs (STS=214 [163-318] vs NAC=122 [67-227]; p<0.001). SLF density was also lower following NAC (0.68 per mm² [0.50-1.01 per mm²]) compared to STS (0.86 per mm² [0.64-1.06 per mm²]; p=0.0354).

Conclusion:
NAC is associated with reduced LN-neg size, SLF count, and SLF density. This validates the findings presented in the previous pilot study in an independent cohort from the same FOxTROT clinical trial. Analysis in the prior cohort had shown better recurrence free survival in patients with greater SLF density in both arms of the trial. Assessment of LN-neg architecture in surgical specimens could be used to guide adjuvant treatment decisions in both the STS and NAC settings.