Esophageal cancer in an adult with congenital esophageal stenosis: a case report.

BACKGROUND: Congenital esophageal stenosis (CES) is a rare condition. We encountered a case of esophageal cancer that developed in an adult with persistent CES. Although many studies have investigated the therapeutic outcomes and performed surveillance for symptoms after treatment for CES, few have performed long-term surveillance or reported on the development of esophageal cancer. We report this case because it is extremely rare and has important implications. CASE PRESENTATION: A 45-year-old woman with worsening dysphagia was transferred to our hospital. The patient was diagnosed with CES at 5 years of age and underwent surgery at another hospital. The patient underwent esophageal dilatation for stenosis at 36 years of age. Esophagoscopy performed at our hospital revealed a circumferential ulcerated lesion and stenosis 15-29 cm from the incisors. Histological examination of the biopsy specimen revealed squamous cell carcinoma. Computed tomography (CT) revealed abnormal circumferential wall thickening in parts of the cervical and almost the entire thoracic esophagus. 18F-fluorodeoxyglucose-positron emission tomography-CT revealed increased uptake in the cervical and upper esophagus. No uptake was observed in the muscular layers of the middle or lower esophagus. Based on these findings, the patient was diagnosed with clinical stage IVB cervical and upper esophageal cancer (T3N1M1 [supraclavicular lymph nodes]). The patient underwent a total esophagectomy after neoadjuvant chemotherapy. The esophagus was markedly thickened and tightly adhered to the adjacent organs. Severe fibrosis was observed around the trachea. Marked thickening of the muscular layer was observed throughout the esophagus; histopathological examination revealed that this thickening was due to increased smooth muscle mass. No cartilage, bronchial epithelium, or glands were observed. The carcinoma extended from the cervical to the middle esophagus, oral to the stenotic region. Finally, we diagnosed the patient with esophageal cancer developing on CES of the fibromuscular thickening type. CONCLUSIONS: Chronic mechanical and chemical irritations are believed to cause cancer of the upper esophagus oral to a persistent CES, suggesting the need for long-term surveillance that focuses on residual stenosis and cancer development in patients with CES.

Vascular anatomical study of persistent descending mesocolon in patients undergoing laparoscopic surgery for colorectal cancer.

INTRODUCTION: Persistent descending mesocolon (PDM) is a rare congenital atypia of fixation of the descending colon, and currently, very few detailed studies exist on its vascular anatomy. This study was conducted to evaluate the features of the vascular anatomy of PDM to help avoid intraoperative lethal injury and subsequent postoperative complications in laparoscopic colorectal surgery. METHODS: We retrospectively analyzed the data of 534 patients who underwent laparoscopic left-sided colorectal surgery. PDM was diagnosed using preoperative axial computed tomography (CT) view. The vascular anatomical features were compared between PDM and non-PDM cases based on three-dimensional (3D)-CT angiography findings. Additionally, the perioperative short-term outcomes of laparoscopic surgery in the 534 patients were also compared between PDM and non-PDM cases. RESULTS: Of the total 534 patients, 13 patients (2.4%) presented with PDM. No branching pattern of the inferior mesenteric artery (IMA) specific to PDM was found. In the running direction of the IMA and sigmoidal colic artery (SA), the midline-shift of IMA and the right-shift of SA were significantly more in PDM than in non-PDM cases, respectively (38.5% vs. 2.5%, P ≤ .0001; 61.5% vs. 4.6%, P ≤ .0001). The perioperative short-term outcomes of laparoscopic surgery in the 534 patients were similar between PDM and non-PDM cases. CONCLUSION: Because changes in the direction of the vascular running are often observed due to adhesions and shortening of the mesentery in PDM cases, performing a detailed preoperative evaluation of vascular anatomy using imaging modalities such as 3D-CT angiography is important.

Optimal bowel resection margin in colon cancer surgery: prospective multicentre cohort study with lymph node and feeding artery mapping.

Background: There are no standardised criteria for the 'regional' pericolic node in colon cancer, which represents a major cause of the international uncertainty regarding the optimal bowel resection margin. This study aimed to determine 'regional' pericolic nodes based on prospective lymph node (LN) mapping. Methods: According to preplanned in vivo measurements of the bowel, the anatomical distributions of the feeding artery and LNs were determined in 2996 stages I-III colon cancer patients who underwent colectomy with resection margin >10 cm at 25 institutions in Japan. Findings: The mean number of retrieved pericolic nodes was 20.9 (standard deviation, 10.8) per patient. In all patients except seven (0.2%), the primary feeding artery was distributed within 10 cm of the primary tumour. The metastatic pericolic node most distant from the primary tumour was within 3 cm in 837 patients, 3-5 cm in 130 patients, 5-7 cm in 39 patients and 7-10 cm in 34 patients. Only four patients (0.1%) had pericolic lymphatic spread beyond 10 cm; all of whom had T3/4 tumours accompanying extensive mesenteric lymphatic spread. The location of metastatic pericolic node did not differ by the feeding artery's distribution. Postoperatively, none of the 2996 patients developed recurrence in the remaining pericolic nodes. Interpretation: The pericolic nodes designated as 'regional' were those located within 10 cm of the primary tumours, which should be fully considered when determining the bowel resection margin, even in the era of complete mesocolic excision. Funding: Japanese Society for Cancer of the Colon and Rectum.

UDP-glucuronosyltransferase 1A1*6 and *28 polymorphisms as indicators of initial dose level of irinotecan to reduce risk of neutropenia in patients receiving FOLFIRI for colorectal cancer.

BACKGROUND: Irinotecan (CPT-11)-induced neutropenia is associated with UDP-glucuronosyltransferase (UGT) 1A1*6 and *28 polymorphisms. This prospective study investigated whether using these polymorphisms to adjust the initial dose of CPT-11 as part of FOLFIRI treatment in colorectal cancer patients might improve safety. METHODS: All data were collected by a physician. The relationship between UGT1A1 polymorphisms and first-cycle neutropenia, reasons for treatment discontinuation, and time-to-treatment failure were evaluated. Multivariate analysis was used to assess the risk of neutropenia. RESULTS: A total of 795 patients were divided into wild-type (*1/*1) (50.1 %), heterozygous (*28/*1, *6/*1) (41.1 %), and homozygous (*28/*28, *6/*6, *28/*6) (8.8 %) groups, in which the median starting dose of CPT-11 was 143.0, 143.0, and 115.0 mg/m(2), respectively. First-cycle grade ≥3 neutropenia occurred in 17.3, 25.4, and 28.6 % of these patients, respectively. Multivariate analysis revealed that the incidence of grade ≥3 neutropenia was significantly greater in the heterozygous and homozygous groups than in the wild-type group [odds ratio (OR) 1.67; 95 % confidence interval (CI) 1.16-2.42; p = 0.0060, and OR 2.22; 95 % CI 1.22-4.02; p = 0.0088, respectively]. Age (OR 1.77; 95 % CI 1.24-2.53; p = 0.0017), coelomic fluid (OR 1.84; 95 % CI 1.05-3.25; p = 0.0343), and non-reduction in starting dose (OR 1.53; 95 % CI 1.08-2.18; p = 0.0176) were also identified as significant risk factors. CONCLUSION: The risk of neutropenia was higher in the heterozygous and homozygous groups at initiation of CPT-11 treatment. This suggests that when a reduction in dose is required in patients harboring two variant alleles, the decrease should be approximately 20 %.

Randomized controlled trial on the skin toxicity of panitumumab in Japanese patients with metastatic colorectal cancer: HGCSG1001 study; J-STEPP.

AIM: We planned a randomized, open-label trial to evaluate differences between pre-emptive and reactive skin treatment for panitumumab (Pmab)-associated skin toxicities in Japanese patients with metastatic colorectal cancer. PATIENTS & METHODS: Patients receiving third-line Pmab-containing regimens were randomized to pre-emptive or reactive treatment. The primary end point was the cumulative incidence of ≥grade 2 skin toxicities during 6 weeks. Retrospectively, a dermatologist reviewed skin toxicities, in a blinded manner. RESULTS: A total of 95 patients were enrolled (pre-emptive: 47, reactive: 48). The primary end point was achieved (21.3 and 62.5% [risk ratio: 0.34; p < 0.001], for pre-emptive and reactive treatment, respectively). A similar trend was observed in central review. CONCLUSION: Pre-emptive skin treatment could reduce the severity of Pmab-associated skin toxicities in Japanese metastatic colorectal cancer patients.

Clinicopathological and molecular characteristics of serrated lesions in Japanese elderly patients.

BACKGROUND: The population in Japan is aging more rapidly than in any other country. However, no studies have determined the characteristics of the large population of elderly patients with colorectal tumors. Therefore, we examined the clinicopathological and molecular features of these tumors in elderly patients. METHODS: In total, 1,627 colorectal tumors (393 serrated lesions, 277 non-serrated adenomas and 957 colorectal cancers) were acquired from patients. Tumor specimens were analyzed for BRAF and KRAS mutations, CpG island methylator phenotype-specific promoters (CACNA1G, CDKN2A, IGF2 and RUNX3), IGFBP7, MGMT, MLH1 and RASSF2 methylation, microsatellite instability (MSI) and microRNA- 31 (miR-31). RESULTS: The frequency of elderly patients (aged ≥75 years) with sessile serrated adenomas (SSAs) with cytological dysplasia was higher than that of those with other serrated lesions and non-serrated adenomas (p < 0.0001). In elderly patients, all SSAs were located in the proximal colon (particularly the cecum to ascending colon). High miR-31 expression, MLH1 methylation and MSI-high status were more frequently detected in SSAs from elderly patients than in those from non-elderly patients. In contrast, no significant differences were found between older age of onset and high-grade dysplasia for traditional serrated adenomas or non-serrated adenomas in any of these molecular alterations. CONCLUSION: In elderly patients, all SSAs were located in the proximal colon. Furthermore, cytological dysplasia and molecular alterations were more frequently detected in elderly patients with SSAs than in non-elderly patients. Thus, careful colonoscopic examinations of the proximal colon are necessary for elderly patients because SSAs in those patients may exhibit malignant potential.

Association of MicroRNA-31-5p with Clinical Efficacy of Anti-EGFR Therapy in Patients with Metastatic Colorectal Cancer.

BACKGROUND: Gene mutations in the pathway downstream of epidermal growth factor receptor (EGFR) are considered to induce resistance to anti-EGFR therapy in colorectal cancer (CRC). We recently reported that microRNA-31 (miR-31)-5p may regulate BRAF activation and play a role in the signaling pathway downstream of EGFR in CRC. Therefore, we hypothesized that miR-31-5p can be a useful biomarker for anti-EGFR therapy in CRC. METHODS: We evaluated miR-31-5p expression and gene mutations [KRAS (codon 61 or 146), NRAS (codon 12, 13, or 61), and BRAF (V600E)] in the EGFR downstream pathway in 102 CRC patients harboring KRAS (codon 12 or 13) wild-type who were treated with anti-EGFR therapeutics. Progression-free survival (PFS) and overall survival (OS) were evaluated. RESULTS: KRAS (codon 61 or 146), NRAS, and BRAF mutations were detected in 6.9, 6.9, and 5.9 % patients, respectively. Compared with CRCs with at least one mutation (n = 20), significantly better PFS (P = 0.0003) but insignificantly better OS were observed in CRCs harboring all wild-type genes (KRAS, NRAS, and BRAF). High miR-31-5p expression was identified in 11 % (n = 11) patients and was significantly associated with shorter PFS (P = 0.003). In CRCs carrying all wild-type genes, high miR-31-5p was associated with shorter PFS (P = 0.027). CONCLUSIONS: High miR-31-5p expression was associated with shorter PFS in patients with CRC treated with anti-EGFR therapeutics. Moreover, in CRCs carrying all wild-type genes, high miR-31-5p was associated with shorter PFS, suggesting that it may be a useful and additional prognostic biomarker for anti-EGFR therapy.

Incisional surgical site infection after elective open surgery for colorectal cancer.

BACKGROUND: The purpose of this study was to clarify the incidence and risk factors for incisional surgical site infections (SSI) in patients undergoing elective open surgery for colorectal cancer. METHODS: We conducted prospective surveillance of incisional SSI after elective colorectal resections performed by a single surgeon for a 1-year period. Variables associated with infection, as identified in the literature, were collected and statistically analyzed for their association with incisional SSI development. RESULTS: A total of 224 patients were identified for evaluation. The mean patient age was 67 years, and 120 (55%) were male. Thirty-three (14.7%) patients were diagnosed with incisional SSI. Multivariate analysis suggested that incisional SSI was independently associated with TNM stages III and IV (odds ratio [OR], 2.4) and intraoperative hypotension (OR, 3.4). CONCLUSIONS: The incidence of incisional SSI in our cohort was well within values generally reported in the literature. Our data suggest the importance of the maintenance of intraoperative normotension to reduce the development of incisional SSI.

Histogenesis and prognostic value of myenteric spread in colorectal cancer: a Japanese multi-institutional study.

BACKGROUND: The histogenesis of the pattern of cancer spread along Auerbach's plexus (myenteric spread: MS) remains unclear and its prognostic value in colorectal cancer (CRC) has not been thoroughly investigated. METHODS: Pathology slides of 2845 pT2/pT3/pT4 CRCs stained with hematoxylin-eosin (H&E) were reviewed at 10 institutions. MS was classified into 2 groups depending on whether it was accompanied by the finding of perineural invasion (PN) within the lesion. In addition, immunohistochemical staining (D2-40, S100, CD56, synaptophysin) was performed for serially sectioned specimens from 50 CRCs diagnosed as having PN-negative MS. RESULTS: MS was observed in 504 patients (17.7 %); 360 patients were classified as having PN-positive MS and 144 as having PN-negative MS. The 5-year disease-free survival rate of patients with MS was lower than that of patients without MS (63.3 vs 82.7 %, P < 0.0001); however, there was no significant difference in survival outcome according to the presence or absence of intralesion PN in MS. Multivariate analysis showed that the prognostic impact of MS was independent of conventional prognosticators including T and N stages, vascular invasion and extramural PN. In all the tumors having PN-negative MS, remnants of neural tissue were identified within or around cancer nests located at the leading edge of MS. CONCLUSIONS: MS is an important prognostic factor for CRC. This feature is the result of cancer development with replacement of Auerbach's plexus and can be classified as intramural PN. The clinical significance of "Pn1" in the UICC/AJCC TNM classification could be enhanced by individual assessment both intramurally and extramurally.

Factors associated with risk for colorectal cancer recurrence after endoscopic resection of T1 tumors.

BACKGROUND & AIMS: More information is needed on the long-term outcomes of patients who undergo endoscopic resection of colorectal tumors. We evaluated recurrence of colorectal cancer (CRC) after endoscopic resection or a combination of endoscopic research and surgery for T1 colorectal tumors. METHODS: We conducted a retrospective study of 389 patients with T1 CRC treated by endoscopic resection from January 1989 to December 2008 in Sapporo, Japan. We compared outcomes between patients who underwent subsequent surgery (ER + SURG, n = 205) and those who did not (ER only, n = 184) and statistically adjusted baseline differences between the groups according to the propensity scores. RESULTS: There was almost no risk of cancer recurrence among patients without indications for surgery recommended by the Japanese Society for Cancer of the Colon and Rectum (these indications include tumors with vertical margins, deep submucosal invasion, lymphatic or venous invasion, poor differentiation, or high-grade budding). Among patients with indications for surgery, the cumulative risks of recurrence (CRRs) were 3.7% in the ER + SURG group and 20.1% in the ER only group (P = .001). However, the patients with only deep submucosal invasion had a low CRR, even without surgery (2.3% in the ER + SURG group and 3.4% in the ER only groups, P = .867). In contrast, patients with indications for surgery other than deep submucosal invasion (high-risk patients) had much better outcomes when they also underwent surgery (CRRs: 5.8% in the ER + SURG group vs 58.0% in the ER only group, P < .001). CONCLUSIONS: On the basis of a retrospective study of patients who underwent endoscopic resection for T1 CRC, those with tumors with only submucosal invasion are at low risk for cancer recurrence. However, patients with other high-risk tumor features have greater risks for cancer recurrence and benefit from subsequent surgery.

Diagnostic value of ¹8F-FDG PET/CT for lymph node metastasis of esophageal squamous cell carcinoma.

PURPOSE: This study investigated the diagnostic accuracy of hybrid positron-emission tomography/computed tomography (PET/CT) for lymph node (LN) metastasis of esophageal cancer. We also investigated the correlation between the size of metastatic nests and the detection by PET/CT. METHODS: Two hundred and fifty-eight patients with esophageal squamous cell carcinoma who underwent esophagectomy with two- or three-field radical lymphadenectomy were analyzed retrospectively. We compared the diagnosis of preoperative PET/CT to the postoperative histopathological examination by each anatomical field (n = 1,231) in all 258 patients. The metastatic LNs resected from PET/CT positive fields were classified as belonging to the PET/CT-N-positive group (n = 229) and those from negative fields as belonging to the PET/CT-N-negative group (n = 352). The cross-sectional areas of metastatic nests were measured in each metastatic LN. RESULTS: Of the 1,231 fields, 275 (22 %) were positive for metastasis, including 581 LNs from 408 regional LN stations. The sensitivity and specificity of PET/CT examined by each anatomical field were 25.8 and 97.8 %, respectively. The median area of metastatic nests was 17.7 mm2 in the PET/CT-N-positive group, and 7.7 mm2 in the PET/CT-N-negative group (p < 0.001). CONCLUSIONS: A significant correlation was suggested between the nest size and detection by PET/CT. Because of its low sensitivity, PET/CT alone is insufficient to determine the surgical procedures, especially when considering reduction surgery.

Association of microRNA-31 with BRAF mutation, colorectal cancer survival and serrated pathway.

BRAF is an important gene in colorectal cancers (CRCs) that is associated with molecular characterization and resistance to targeted therapy. Although microRNAs (miRNAs) are useful biomarkers of various cancers, the association between miRNA and BRAF in CRCs is undefined. Therefore, this study was conducted to identify a relationship between specific miRNA molecules and BRAF mutation in CRCs and serrated lesions. miRNA array was used for the measurement of 760 miRNAs in 29 CRCs. To assess the identified miRNAs, quantitative reverse transcription-PCR was performed on 721 CRCs, 381 serrated lesions and 251 non-serrated adenomas. Moreover, proliferation and invasion assays were conducted using cell lines. miRNA array analysis revealed that microRNA-31 (miR-31)-5p was the most up-regulated miRNA in CRCs with mutated BRAF (V600E) compared with CRCs possessing wild-type BRAF (including cases with KRAS mutation). High miR-31 expression was associated with BRAF and KRAS mutations and proximal location (P < 0.0001). High miR-31 expression was related to cancer-specific mortality [multivariate hazard ratio = 2.06, 95% confidence interval: 1.36-3.09, P = 0.0008]. Functional analysis demonstrated that miR-31 inhibitor decreased cell invasion and proliferation. With regard to serrated lesions, high miR-31 expression was less frequently detected in hyperplastic polyps compared with other serrated lesions. In conclusion, associations were identified between miR-31, BRAF and prognosis in CRC. Transfection of miR-31 inhibitor had an antitumour effect. Thus, miR-31 may be a promising diagnostic biomarker and therapeutic target in colon cancers. Moreover, high miR-31 expression in serrated lesions suggested that miR-31 may be a key molecule in serrated pathway.

Characterization of perineural invasion as a component of colorectal cancer staging.

Perineural invasion (PN) in colorectal cancer (CRC) is a site-specific prognostic marker, as mentioned by the AJCC Cancer Staging Manual, but it remains to be clearly defined. We aimed to identify an optimal characterization of PN as a component of cancer staging. On the basis of the anatomic features of the nervous system of the large bowel, site-specific pathologic criteria were assigned to PN according to the location of PN. Multi-institutional pathologic review based on these criteria was performed for 962 patients with stage I to III CRC at 2 institutions (1999 to 2004, cohort 1) and 1883 patients from 8 other institutions (2000 to 2004, cohort 2). In cohort 1, intramural and extramural PN were observed in 152 and 101 patients, respectively, which had a different impact on disease-free survival (hazard ratio, 2.6 [1.9 to 3.5] vs. 4.7 [3.4 to 6.5], respectively). A 3-tiered grading system (Pn0; Pn1a, intramural PN; Pn1b, extramural PN) distinguished 5-year disease-free survival as 88%, 70%, and 48%, respectively; and multivariate analysis identified PN grade as a significant prognostic marker independent of T or N stage. These results were similar in cohort 2. Interinstitutional difference of the prognostic impact of PN grade was acceptably small among all institutions. Interobserver study among 6 gastrointestinal pathologists showed superior judgment reproducibility for PN compared with vascular invasion. The results of our study indicate that PN is an important prognostic marker in CRC. The value of cancer staging could be enhanced by PN assessment using site-specific criteria and a simple grading system based on PN location within the bowel.

Optimal colorectal cancer staging criteria in TNM classification.

PURPOSE: Histologic components of the TNM classification system have been repeatedly revised since the fifth edition (TNM5). TNM classification revisions provide different criteria for categorizing tumor nodules without residual lymph node structure (ND). However, there are few systematic evaluations regarding the effectiveness of these revisions. PATIENTS AND METHODS: A multicenter pathologic review for ND in colorectal cancer (CRC) was performed. Tumor staging defined by TNM5, sixth edition (TNM6), and seventh edition (TNM7) were compared on the basis of Akaike information criterion (AIC) and Harrell's concordance index (c-index). Moreover, TNM7's prognostic value was compared between the original ND and modified criteria, which considered all regional NDs as lymph node metastasis (LNM) irrespective of the original structure. RESULTS: In 1,716 treated patients with CRC (1994 to 1998), tumor stages (I/II/III) exhibited better prognoses in TNM7 (AIC, 3055.1; c-index, 0.7215) than in TNM6 (AIC, 3063.7; c-index, 0.7149), but not better than in TNM5 (AIC, 3051.6; c-index, 0.7240). Comparing the original TNM7 and modified criteria, 4.2% of patients were classified in different N stages (N0/N1/N2a/N2b); both AIC and the c-index were superior in the modified criteria (AIC, 3029.40; c-index, 0.7271) compared with the original criteria (AIC, 3040.58; c-index, 0.7230). Modified criteria were also associated with improved prognostic power of tumor stages (I/IIA/IIB/IIC/IIIA/IIIB/IIIC). These results were similar in another cohort of 2,242 treated patients with CRC (1999 to 2003). CONCLUSION: The prognostic value of TNM7 is better than that of TNM6; however, improvement over TNM5 is insignificant. By considering all regional NDs as LNM irrespective of their morphology, TNM classification can be simplified and its prognostic value improved.

Multicenter study for optimal categorization of extramural tumor deposits for colorectal cancer staging.

OBJECTIVE: This study aimed to determine the optimal categorization of extramural tumor deposits lacking residual lymph node (LN) structure (EX) in colorectal cancer staging. BACKGROUND: The TNM classification system categorizes EX on the basis of their contour characteristics (the contour rule). METHODS: We conducted a multicenter, retrospective, pathological review of 1716 patients with stage I to III curatively resected colorectal cancer who were treated at 11 institutions (1994-1998). In addition, 2242 patients from 9 institutions (1999-2003) were enrolled as a second cohort for validating results. EX were classified as isolated foci confined to vascular or perineural spaces (ie, lymphatic, venous, or perineural invasion) or as tumor nodules (ND). N- and T-staging systems employing different categories for staging were compared in terms of their prognostic power. In addition, the diagnoses of extramural, discontinuously spreading lesions made by 11 observers from different institutions were assessed for interobserver agreement. RESULTS: EX were observed in 18.2% of patients in the first cohort. The method of categorization of EX in tumor staging has a stronger impact on N than T staging. The N-staging system in which all ND types were classified as N factor (the ND rule) could more effectively stratify the survival outcome than the contour rule (Akaike information criterion, 3040.8 vs 3059.5; the Harrell C-index, 0.7255 vs 0.7103). EX were observed in 16.9% of patients in the second cohort. Statistically, the ND rule was more informative than the contour rule for N staging. The Fleiss kappa coefficient for distinguishing LN metastases from EX (0.74) was lower than expected for complete agreement, and it decreased further to 0.51 when calculated for the judgment of ND with smooth contours. CONCLUSIONS: Classifying all ND types as N factors irrespective of contours can simplify the tumor staging system by enhancing diagnostic objectivity, resulting in improved prognostic accuracy.

Actual status of distribution and prognostic impact of extramural discontinuous cancer spread in colorectal cancer.

PURPOSE: To clarify the prognostic impact of tumor nodules without residual lymph node (LN) structure (ND) in colorectal cancer and to determine optimal categorization of ND in tumor staging. PATIENTS AND METHODS: A multicenter, retrospective pathologic review was performed for 1716 patients with stages I to III curatively resected colorectal cancer treated at 11 institutions between 1994 and 1998. An additional 2242 patients from nine institutions were enrolled between 1999 and 2003 as a second cohort to validate the results. RESULTS: LN metastasis (LNM) and ND were observed in 33.7% and 16.0% (smooth-contour nodule [S-ND], 8.2%; irregular-contour nodule [I-ND], 10.7%) of patients in the first cohort. S-ND and I-ND were similarly distributed in the regional area. There was no considerable difference in the impact on survival between S-ND (hazard ratio [HR], 2.7; 95% CI, 1.9 to 3.8) and I-ND (HR, 4.3; 95% CI, 3.3 to 5.8) or between LNM (HR, 4.5; 95% CI, 3.4 to 6.0) and ND (HR, 4.0; 95% CI, 3.1 to 5.3). LNM and ND were similarly associated with the mode of recurrence. Tumor nodules ≥ 5 mm growing with venous/perineural invasion (ND [v/pni+]), judged with 0.61 κ value among 11 observers, had an independent prognostic value for 5-year survival of 42%; similar results were observed in the second cohort. CONCLUSION: These results do not support the TNM system in which S-ND is treated differently from I-ND in tumor staging; LNM and ND should be considered together in the same category. The presence of ND (v/pni+) has a considerable adverse prognostic effect.

Retrospective cohort study on the safety and efficacy of bevacizumab with chemotherapy for metastatic colorectal cancer patients: the HGCSG0801 study.

OBJECTIVE: After approval of bevacizumab in Japan, post-marketing surveillance studies reported on safety. However, few reports have shown the efficacy of bevacizumab as used in daily practice. We evaluated the efficacy and safety of bevacizumab for metastatic colorectal cancer patients in daily practice. METHODS: All unresectable metastatic colorectal cancer patients who began receiving bevacizumab in participating facilities from June 2007 to October 2008 were retrospectively analyzed for safety and efficacy. Adverse events were assessed by the National Cancer Institute Common Terminology Criteria for Adverse Events. Response Evaluation in Solid Tumors criteria, version 1.0, was used for the tumor response rate. RESULTS: A total of 212 patients from 17 institutions were assessed. Grade 3 or higher adverse events related to bevacizumab included gastrointestinal perforation in 3, thrombosis in 7, hypertension in 30 and gastrointestinal bleeding in 2. Response rates were 62.5, 30.1 and 11.8% overall among patients receiving bevacizumab as first-, second- and third-line or greater therapy. Median progression-free survival was 14.4 [95% confidence interval (CI): 10.8-18.1], 7.8 (95% CI: 6.5-9.1) and 6.0 (95% CI: 4.6-7.3) months, and median overall survival was 32.5 (95% CI: 24.6-40.3), 16.4 (95% CI: 14.4-18.5) and 11.8 (95% CI: 8.6-15.0) months, respectively. CONCLUSIONS: The general cohort of patients in HGCSG0801 showed a similar efficacy and safety profile of bevacizumab as seen in clinical trials. Although the sample size was small and there were several study limitations, these results suggest that colorectal cancer patients in Japan might safely receive and benefit from bevacizumab in combination with chemotherapy in daily practice, as is seen in patients in other countries.

Phase I/II study of S-1 combined with carboplatin in recurrent and/or metastatic head and neck cancer as outpatient chemotherapy.

BACKGROUND: The aims of this study were to determine the recommended dose (RD) in combination chemotherapy of S-1 and carboplatin in recurrent and/or metastatic head and neck cancer (phase I) and to examine response rate and toxicities at RD (phase II). METHODS: S-1 was given orally at a fixed dose of 40, 50, or 60 mg twice daily based on the patient's body surface area for 21 days followed by a 14-day rest. In phase I, carboplatin was given intravenously to a cohort of three patients on day 8 at a dose of 2.5, 3.0, 3.5, or 4.0 area under the curve (AUC) values, depending on the dose-limiting toxicities (DLTs). RESULTS: In phase I, the RD was estimated as 2.5 AUC. In phase II, thrombocytopenia and leukopenia were most commonly found as hematologic DLTs, which were manageable without hospitalization. The response rate was 40.9%. CONCLUSIONS: This regimen is considered to be active against recurrent and/or metastatic head and neck cancer in an outpatient setting.

The long-term voiding function and sexual function after pelvic nerve-sparing radical surgery for rectal cancer.

BACKGROUND: The aim of the present study is to symptomatically analyze the extent to which pelvic nerve-sparing radical surgery for rectal cancer impacts on long-term voiding and male sexual function. METHODS: A self-administered questionnaire was mailed to 68 patients who underwent pelvic nerve-sparing radical surgery for invasive rectal cancer with 52 responses (28 men and 24 women; 27 complete and 25 incomplete preservation; response rate 76.5%). Each patient was asked to record if there had been any changes in lower urinary tract symptoms after surgery. Sexual function was also investigated in men. RESULTS: Of the 52 patients, 48 (92%) maintained voluntary voiding without catheterization in the long term. Clean intermittent self-catheterization was performed in only four patients with incomplete preservation because of persistent voiding dysfunction. Subjectively, approximately 60% of the patients remained unchanged in lower urinary tract symptoms after surgery. The satisfaction rate regarding the current voiding status was significantly higher in women than in men (83% versus 61%, P = 0.0294), but was not significantly different between those with complete (76%) and incomplete preservation (64%). Despite the acceptable urinary status, 88% of men had some deterioration in the erectile function, regardless of the types of surgical procedures. Overall, 64% of men were unsatisfied with the current sexual function. CONCLUSIONS: Pelvic nerve-sparing radical surgery for rectal cancer preserved the long-term voiding function in the majority of patients. In completely preserved patients and in women, symptomatic outcomes were more satisfactory. Postoperative erectile dysfunction was found to be a serious problem, even in complete nerve-sparing procedure.