Journal of Surgical Oncology最新文献

Background and objectives: The combination of focused ultrasound (FUS) and chemotherapy is a novel treatment for pancreatic cancer. This paper reviews the literature on this combined therapy.

Methods: The medical literature was searched according to PRISMA guidelines. Inclusion criteria were any study of patients with pancreatic cancer undergoing treatment with FUS and chemotherapy. Data extracted included stage, radiologic response, resection rate, survival, and adverse events.

Results: The initial search yielded 212 citations; 10 studies met inclusion criteria (9 retrospective cohorts; 1 randomized trial). A total of 631 patients received FUS + chemotherapy; 63.6% being stage 4, and 29.7% stage 3. Patient selection, FUS parameters, and chemotherapy used were all heterogeneous. Overall survival ranged from 7.4 to 21.6 months, radiologic response rate was 44.1%, and 24.4% of stage 3 patients underwent resection. All four studies with a comparison group demonstrated improved survival. FUS + chemotherapy decreased pain in 69.7% of patients. Severe adverse events occurred in 0.65%.

Conclusions: The literature on combined FUS and chemotherapy for pancreatic cancer is heterogeneous. There is good evidence that adverse events are low, and that it provides effective palliation. There is evidence to suggest oncologic benefit, however, this is subject to selection bias and prospective trials are needed.

Background: In patients with localized pancreatic ductal adenocarcinoma (PDAC) undergoing neoadjuvant therapy (NAT) and resection, selection of adjuvant chemotherapy (AC) is typically guided by high-risk features on histopathologic examination. We evaluated the interaction between post-NAT lymph node metrics and AC receipt on survival.

Methods: Patients who received NAT followed by pancreatectomy (2010-2020) at seven centers were reviewed. Overall survival (OS) in patients receiving AC or not was stratified by lymph node positivity (LNP) or lymph node ratio (LNR) dichotomized at 0.1. Cox models evaluated the independent association between these nodal metrics, AC receipt, and OS.

Results: Of 464 patients undergoing NAT and resection, 264 (57%) received AC. Patients selected for AC were younger (median 63 vs. 67 years; p < 0.001), received shorter duration of NAT (2.8 vs. 3.2 months; p = 0.01), had fewer postoperative complications (Clavien-Dindo grade > 3: 1.2% vs. 11.7%; p < 0.001), and lower rates of pathologic complete response (4% vs. 11%; p = 0.01). The median number of nodes evaluated was similar between cohorts (n = 20 in both; p = 0.9). Post-NAT LNP rates were not different, and median LNR was 0.1, in AC and non-AC cohorts. Both LNP (hazard ratio [HR]: 2.1, p < 0.001) and LNR (0 < LNR ≤ 0.1: HR: 1.98, p = 0.002; LNR > 0.1: HR 2.46, p < 0.001) were independently associated with OS on Cox modeling, although receipt of AC was not associated with improved OS (median 30.6 vs. 29.4 months; p = 0.2). In patients with LNR > 0.1, receipt of AC was associated with significantly longer OS compared to non-AC (24 vs. 20 months, respectively; p = 0.04).

Conclusions: LNR following NAT, not simply nodal positivity, may be useful to refine selection of AC in resected PDAC.

Background: Chemotherapy enhances survival rates for pancreatic cancer (PC) patients postsurgery, yet less than 60% complete adjuvant therapy, with a smaller fraction undergoing neoadjuvant treatment. Our study aimed to predict which patients would complete pre- or postoperative chemotherapy through machine learning (ML).

Methods: Patients with resectable PC identified in our institutional pancreas database were grouped into two categories: those who completed all intended treatments (i.e., surgery plus either neoadjuvant or adjuvant chemotherapy), and those who did not. We applied logistic regression with lasso penalization and an extreme gradient boosting model for prediction, and further examined it through bootstrapping for sensitivity.

Results: Among 208 patients, the median age was 69, with 49.5% female and 62% white participants. Most had an Eastern Cooperative Oncology Group (ECOG) performance status of ≤2. The PC predominantly affected the pancreatic head. Neoadjuvant and adjuvant chemotherapies were received by 26% and 47.1%, respectively, but only 49% completed all treatments. Incomplete therapy was correlated with older age and lower ECOG status. Negative prognostic factors included worsening diabetes, age, congestive heart failure, high body mass index, family history of PC, initial bilirubin levels, and tumor location in the pancreatic head. The models also flagged other factors, such as jaundice and specific cancer markers, impacting treatment completion. The predictive accuracy (area under the receiver operating characteristic curve) was 0.67 for both models, with performance expected to improve with larger datasets.

Conclusions: Our findings underscore the potential of ML to forecast PC treatment completion, highlighting the importance of specific preoperative factors. Increasing data volumes may enhance predictive accuracy, offering valuable insights for personalized patient strategies.

Background: In retroperitoneal leiomyosarcoma (RP LMS), the predominant issue is distant metastasis (DM). We sought to determine variables associated with this outcome and disease-specific death (DSD).

Methods: Data were retrospectively collected on patients with primary RP LMS treated at a high-volume center from 2002 to 2023. For inferior vena cava (IVC)-origin tumors, the extent of macroscopic vascular invasion was re-assessed on each resection specimen and correlated with preoperative cross-sectional imaging. Crude cumulative incidences were estimated for DM and DSD and univariable and multivariable models were performed.

Results: Among 157 study patients, median tumor size was 11.0 cm and 96.2% of cases were intermediate or high grade. All patients underwent complete resection, 56.7% received chemotherapy (43.9% neoadjuvant) and 14.6% received radiation therapy. Only tumor size and grade and not site of tumor origin (e.g., IVC vs. other) were associated with DM and DSD (p < 0.05). Among 64 patients with IVC-origin tumors, a novel 3-tier classification was devised based on the level of intimal disruption, which was associated with both DM (p = 0.007) and DSD (0.002).

Conclusion: In primary RP LMS, only tumor size and grade are predictive of DM and DSD. In IVC-origin tumors, the extent of macroscopic vascular invasion is also strongly predictive of these outcomes.

Background and objectives: Our aim in this study was to investigate the usefulness of circulating tumor (ct) DNA methylation analysis for predicting long-term outcomes after resection in Stage IV colorectal cancer (CRC).

Methods: Methylation analyses were performed on 95 plasma samples from patients with CRC who underwent surgery. The methylation status (relative methylation value: RMV) of CpG within the promoter region of three genes (CHFR, SOX11, and CDO1) was assessed to quantitative methylation-specific PCR (qMSP) analysis.

Results: In the patients who had undergone resection of the primary tumor and metastatic organs with curative intent, the CHFR-RMV high group had significantly worse recurrence-free survival (RFS) compared with the CHFR-RMV low group (p = 0.001). Multivariate analysis revealed that CHFR-RMV was a significant independent prognostic factor (hazard ratio = 2.63 (1.29-5.36); p = 0.008). In the patients who had undergone resection of the primary tumor with metastatic organs with curative intent after neoadjuvant systemic chemotherapy, the SOX11-RMV high group had significantly worse RFS compared with the SOX11-RMV low group (p = 0.004).

Conclusions: The current study showed the usefulness of ctDNA methylation analysis for predicting the possibility of curative resection and long-term outcomes after resection in Stage IV CRC. A future prospective study is needed to obtain more conclusive results.

Secondary liver malignancies are a serious and challenging global health concern. Secondary metastasis to the liver is most commonly from colorectal cancer that has metastatically spread through splanchnic circulation. Metastatic diseases can portend poor prognosis due to the progressive nature typically found on detection. Improvements in detection of disease, monitoring therapy response, and monitoring for recurrence are crucial to the improvement in the management of secondary liver malignancies. Assessment of ctDNA in these patient populations poses an opportunity to impact the management of secondary liver malignancies. In this review, we aim to discuss ctDNA, the current literature, and future directions of this technology within secondary liver malignancies.

Background: Previous studies report promising outcomes with minimally invasive (MIS) hepatectomy in elderly patients but remain limited by small size. This study aims to comparatively evaluate the demographics and outcomes of geriatric patients undergoing MIS and open hepatectomy.

Method: The 2016-2021 NSQIP database was evaluated comparing patients ≥75 undergoing MIS versus open hepatectomy. Patient selection and outcomes were compared using bivariate analysis with multivariable modeling (MVR) evaluating factors associated with serious complications and mortality. Propensity score matched (PSM) analysis further evaluated serious complications, mortality, length of stay (LOS), Clavien Dindo Classification (CDC), and Comprehensive Complication Index (CCI) for cohorts.

Results: We evaluated 2674 patients with 681 (25.5%) receiving MIS hepatectomy. MIS approaches were used more for partial lobectomy (85.9% vs. 61.7%; p < 0.001), and required fewer biliary reconstructions (1.6% vs. 10.6%; p < 0.001). Patients were similar with regards to sex, body mass index, and other comorbidities. Unadjusted analysis demonstrated that MIS approaches had fewer serious complications (8.8% vs. 18.7%; p < 0.001). However, after controlling for cohort differences the MIS approach was not associated with reduced likelihood of serious complications (odds ratio [OR]: 0.77; p = 0.219) or mortality (OR: 1.19; p = 0.623). PSM analysis further supported no difference in serious complications (p = 0.403) or mortality (p = 0.446). However, following PSM a significant reduction in LOS (-1.99 days; p < 0.001), CDC (-0.26 points; p = 0.016) and CCI (-2.79 points; p = 0.022) was demonstrated with MIS approaches.

Conclusions: This is the largest study comparing MIS and open hepatectomy in elderly patients. Results temper previously reported outcomes but support reduced LOS and complications with MIS approaches.

It remains unclear whether the application of surgery to gastric cancer with peritoneal carcinomatosis prolongs survival. Twenty studies on conversion surgery were reviewed. Key points were the response to chemotherapy, complete resection, and a low tumor burden at the time of surgery. A bidirectional approach has been developed to increase the response rate. There are two different strategies in surgery. The outcomes of ongoing trials may clarify controversial issues.

Background: Large Language Models (LLM; e.g., ChatGPT) may be used to assist clinicians and form the basis of future clinical decision support (CDS) for colon cancer. The objectives of this study were to (1) evaluate the response accuracy of two LLM-powered interfaces in identifying guideline-based care in simulated clinical scenarios and (2) define response variation between and within LLMs.

Methods: Clinical scenarios with "next steps in management" queries were developed based on National Comprehensive Cancer Network guidelines. Prompts were entered into OpenAI ChatGPT and Microsoft Copilot in independent sessions, yielding four responses per scenario. Responses were compared to clinician-developed responses and assessed for accuracy, consistency, and verbosity.

Results: Across 108 responses to 27 prompts, both platforms yielded completely correct responses to 36% of scenarios (n = 39). For ChatGPT, 39% (n = 21) were missing information and 24% (n = 14) contained inaccurate/misleading information. Copilot performed similarly, with 37% (n = 20) having missing information and 28% (n = 15) containing inaccurate/misleading information (p = 0.96). Clinician responses were significantly shorter (34 ± 15.5 words) than both ChatGPT (251 ± 86 words) and Copilot (271 ± 67 words; both p < 0.01).

Conclusions: Publicly available LLM applications often provide verbose responses with vague or inaccurate information regarding colon cancer management. Significant optimization is required before use in formal CDS.

Background: Pancreatic neuroendocrine tumors (pNETs) are genomically diverse tumors. The management of newly diagnosed well-differentiated pNETs is limited by a lack of sensitivity of existing biomarkers for prognostication. Our goal was to investigate the potential utility of genetic markers as a predictor of progression-free survival (PFS) and recurrence-free survival (RFS).

Methods: Whole-exome sequencing of resected well-differentiated, low and intermediate-grade (G1 and G2) pNETs and normal adjacent tissue from patients who underwent resection from 2005 to 2015 was performed. Genetic alterations were classified using pan-genomic and oncogenic pathway classifications. Additional samples with genetic and clinicopathologic data available were obtained from the publicly available International Cancer Genome Consortium (ICGC) database and included in the analysis. The prognostic relevance of these genomic signatures on PFS and RFS was analyzed.

Results: Thirty-one patients who underwent resection for pNET were identified. Genomic analysis of mutational, copy number, cytogenetic, and complex phenomena revealed similar patterns to prior studies of pNETs with relatively few somatic gene mutations but numerous instances of copy number changes. Analysis of genomic and clinicopathologic outcomes using the combined data from our study as well as the ICGC pNET cohort (n = 124 patients) revealed that the recurrent pattern of whole chromosome loss (RPCL) and metastatic disease were independently associated with disease progression. When evaluating patients with local disease at the time of resection, RPCL and alterations in the TGFβ oncogenic pathway were independently associated with the risk of recurrence.

Conclusions: Well-differentiated pNETs are genomically diverse tumors. Pathway signatures may be prognostic for predicting disease progression and recurrence.