Annals of Surgical Oncology

Cell-cycle protein (p27, p21, and p53) expression can predict response to neoadjuvant chemotherapy and prognosis in some neoplasms. This study evaluated whether these markers could also be effective in invasive thymoma during a multimodality treatment. Between 1989 and 2008, 33 patients with invasive thymoma underwent surgical resection after neoadjuvant chemotherapy. Expression of p27, p21, and p53 was assessed using immunohistochemistry in specimens retrieved pre and post chemotherapy. Factors influencing response to neoadjuvant chemotherapy and survival were investigated by univariate and multivariate analysis. Good response was defined as complete disappearance of tumor at imaging or necrosis >90% at pathologic studies. Twelve patients disclosed an imaging good response. Complete resection was possible in 17 patients, 9 of whom had presented imaging good response and 11 of whom had revealed pathologic good response. On univariate analysis both imaging and pathologic poor responses were significantly associated with incomplete resection (P = 0.04 and P = 0.03, respectively) and preneoadjuvant triple combination of p27 low, p21 low, and p53 high expressions (P = 0.001 and P < 0.0001, respectively), the last factor being the only one selected on logistic regression (P = 0.01 and P = 0.005, respectively). Long-term survival analysis was negatively influenced by triple combination of p27, p21, and p53 (P < 0.0001) and incomplete resection (P < 0.0001), which were also selected on Cox’s regression (P = 0.004 and P = 0.02, respectively). The triple combination of p27 low, p21 low, and p53 high expressions was the most significant predictor of imaging and pathologic poor responses to neoadjuvant chemotherapy in invasive thymoma. This combination together with incomplete resection was also the most significant negative predictor of long-term survival.

We focused on the impact of postoperative infection on long-term survival after potentially curative resection for gastric cancer. Postoperative surgical and medical complications have been implicated as a negative predictor of long-term outcome in various malignancies. However, there have been no published reports assessing the impact of complications arising from postoperative infection on survival in gastric cancer. We studied a population of 1,332 patients who underwent curative resection for gastric cancer. These patients were divided into two groups based on the occurrence (141 patients, 10.6%) or absence (1,191 patients, 89.4%) of postoperative complications due to infection. We investigated the demographic and clinicopathological features of each patient with and without postoperative complications from infection, and thereby the impact of postoperative infection on long-term survival. Patients with postoperative infection had significantly higher frequency of males, upper side tumor location, and total gastrectomy as a surgical procedure, more advanced stage of gastric cancer, and greater age compared with those without postoperative infection. Patients with complications due to postoperative infection had significantly more unfavorable outcome compared with those patients without postoperative infection. Multivariate analysis demonstrated that age, preoperative comorbidity, blood transfusion, tumor depth, nodal involvement, and postoperative infection correlated with overall survival. We conclude that postoperative complications from infection are a predictor of adverse clinical outcome in patients with gastric cancer. However, further immunological study and prospective trials are necessary to confirm the biological significance of these findings.

Lymph node metastasis is the most important prognostic indicator for colon cancer patients. We compared the prognostic significance of the number of lymph node metastases (LNN) and the distribution of lymph node metastases (LND). A total of 187 patients underwent curative resection for stage III right-sided colon cancer between 2000 and 2010. We evaluated the oncologic outcomes according to LNN (N1 1–3, N2 4–6, N3 >6) and LND (LND1 metastases in pericolic nodes, LND2 metastases along the major vessels, N3 metastases around the origin of a main artery). A Cox proportional hazards model, with backward stepwise analysis was used to determine the effects of covariates on 5-year, disease-free survival (DFS) and 5-year overall survival (OS). Akaike’s information criterion (AIC), and Harrell’s concordance index (C-index) were compared for each developed model. During the median follow-up of 42.2 months, 5-year DFS and OS were 68 and 79.3 %, respectively. Multivariate analysis showed that both LNN and LND3 were independent prognostic factor for both 5-year DFS and OS. However, the prognostic model incorporating number of LNM was more precise than that of LND, with a lower AIC (5-year DFS, 554.2 vs. 566.9; 5-year OS, 318.1 vs. 337.9) and higher C-index (5-year DFS, 0.706 vs. 0.667; 5-year OS, 0.778 vs. 0.743). Our results show that the staging system incorporating LNN predicted prognosis better than LND.

Several studies suggest that young patients may derive less oncologic benefit from surgical resection of cancers compared with older patients. We hypothesized that young patients may have worse outcomes following cytoreductive surgery with hyperthermic intraperitoneal chemoperfusion (CRS/HIPEC) for peritoneal metastases. Perioperative and oncologic outcomes in adolescent and young adults (AYA), defined as younger than age 40 years (n = 135), undergoing CRS/HIPEC between 2001 and 2015 were reviewed and compared with middle-aged adults, defined as aged 40–65 years (n = 684). The two groups were similar with regards to perioperative characteristics except that AYA were more likely to be symptomatic at presentation (65.2 vs. 50.9%, p = 0.003), had lower Charleson comorbidity index (median 6 vs. 8, p < 0.001), were less likely to receive neoadjuvant chemotherapy (32.8 vs. 42.5%, p = 0.042), and had longer operative times (median 543 vs. 493 min, p = 0.010). Postoperative Clavien–Dindo grade 3–4 morbidity was lower in AYA (17 vs. 26%, p = 0.029), and they required fewer reoperations for complications (3.7 vs. 10.4%, p = 0.014). AYA had longer median overall survival (103.6 vs. 73.2 months, p = 0.053). In a multivariate Cox regression analysis, age was an independent predictor of improved overall survival [hazard ratio 0.705; 0.516–0.963, p = 0.028]. Young patients with peritoneal metastases derive similar benefits from CRS/HIPEC as middle-aged patients. Young age should not be a deterrent to consideration of CRS/HIPEC for peritoneal metastases.

Sentinel lymph node biopsy (SNB) in pregnant women with breast cancer is uncommonly pursued given concern for fetal harm. This study evaluated efficacy and safety outcomes in pregnant breast cancer patients undergoing SNB. Patients who underwent SNB while pregnant were identified from a retrospective parent cohort of women diagnosed with breast cancer during pregnancy. Chart review was performed to tabulate patient/tumor characteristics, method/outcome of SNB, and short-term maternal/fetal outcomes. Within a cohort of 81, 47 clinically node-negative patients had surgery while pregnant: 25 (53.2 %) SNB, 20 (42.6 %) upfront axillary lymph node dissection, and 2 (4.3 %) no lymph node surgery. Of SNB patients, 8, 9, and 8 had SNB in the first, second, and third trimesters, respectively. 99 m-Technetium (99-Tc) alone was used in 16 patients, methylene blue dye alone in 7 patients, and 2 patients had unknown mapping method. Mapping was successful in all patients. There were no SNB-associated complications. At a median of 2.5 years from diagnosis, there was one locoregional recurrence, one new primary contralateral tumor, three distant recurrences, and one breast cancer death. Among patients who underwent SNB, there were 25 liveborn infants, of whom 24 were healthy, and 1 had cleft palate (in the setting of other maternal risk factors). SNB in pregnant breast cancer patients appears to be safe and accurate using either methylene blue or 99-Tc. This is one of the largest reported experiences of SNB during pregnancy; however, numbers remain limited. SNB rates in this cohort were lower than in non-pregnant breast cancer patients.

The aim of this study was to determine the impact of the results of the 12-gene DCIS Score assay on (i) radiotherapy recommendations for patients with pure ductal carcinoma in situ (DCIS) following breast-conserving surgery (BCS), and (ii) patient decisional conflict and state anxiety. Thirteen sites across the US enrolled patients (March 2014–August 2015) with pure DCIS undergoing BCS. Prospectively collected data included clinicopathologic factors, physician estimates of local recurrence risk, DCIS Score results, and pre-/post-assay radiotherapy recommendations for each patient made by a surgeon and a radiation oncologist. Patients completed pre-/post-assay decisional conflict scale and state-trait anxiety inventory instruments. The analysis cohort included 127 patients: median age 60 years, 80 % postmenopausal, median size 8 mm (39 % ≤5 mm), 70 % grade 1/2, 88 % estrogen receptor-positive, 75 % progesterone receptor-positive, 54 % with comedo necrosis, and 18 % multifocal. Sixty-six percent of patients had low DCIS Score results, 20 % had intermediate DCIS Score results, and 14 % had high DCIS Score results; the median result was 21 (range 0–84). Pre-assay, surgeons and radiation oncologists recommended radiotherapy for 70.9 and 72.4 % of patients, respectively. Post-assay, 26.4 % of overall recommendations changed, including 30.7 and 22.0 % of recommendations by surgeons and radiation oncologists, respectively. Among patients with confirmed completed questionnaires (n = 32), decision conflict (p = 0.004) and state anxiety (p = 0.042) decreased significantly from pre- to post-assay. Individualized risk estimates from the DCIS Score assay provide valuable information to physicians and patients. Post-assay, in response to DCIS Score results, surgeons changed treatment recommendations more often than radiation oncologists. Further investigation is needed to better understand how such treatment changes may affect clinical outcomes.