Eric Pihl

Lung recurrence after curative surgery for colorectal cancer

A total of 1578 patients were treated with potentially curative surgical resection for colon and rectal cancer by one surgeon from 1950 to 1982. Follow-up revealed that 117 (11,5 percent) of 1013 patients with rectal carcinoma eventually presented with clinical evidence of pulmonary recurrence, with or without evidence of spread elsewhere; the corresponding figures for the colon were 20 (3.5 percent) of 565 (P<0.001). An analysis of the times to recurrence revealed that half of the lung recurrence, were clinically obvious within 32 months for rectal tumors and 34 months for colonic, compared to 22 and 21 months, respectively, for liver recurrences, excluding those with other distant metastases. The slower recurrence rate and the longer survival in patients with recurrences in the lung compared to the liver were statistically significant only for rectal primaties (P<0.02 andP=0.001, respectively). Sixteen patients underwent surgeery with curative intention for lung recurrences; four of these remain alive at two, six, 11, and 15 years, and one patient was free of recurrence when he died from other causes 15 months after surgery. The conditional probability survival rate for the 16 patients was 38±13 percent at five years after recurrence operation.

I. Carcinoma of the rectum and rectosigmoid: Cancer specific long-term survival. A series of 1061 cases treated by one surgeon

The long‐term cancer specific survival based on individual follow‐up and analysis of prospectively collected data from 1061 patients undergoing resection for carcinoma of the rectum is presented. All patients were operated on and managed by one surgeon. Survival data for 978 cases were analyzed according to the methods of Kaplan and Meier,9 and Gehan.6 Results have been presented as cancer specific survival times in months and as percentage survivor rates at five‐year intervals. The median overall cancer specific survival time was 96 months. Five, ten, 15, and 20‐year survival rates were 56, 49, 47, and 46%, respectively. After a potentially curative resection of the tumor, the corresponding percentages were 69, 60, 57, and 56%. Age and sex were not significant prognostic factors. A death rate from recurrent cancer of nil was seen after 15.4 years. At this point, the cancer specific survival rates were 77% for patients in Stage A, 59% in Stage B, 37% in Stage C, and 9% for patients with tumors invading adjacent organs (D1), while no patient with macroscopic metastases to distant organs (Stage D2) survived beyond four and a half years (median, 11 months).

Mucinous colorectal carcinoma: Immunopathology and prognosis

&NA; A total of 519 colorectal carcinomas were examined for the presence or absence of mucinous differentiation by means of microscopical morphometry. Of these, 28% had objectively measurable amounts of mucinous tumour epithelium. Tumours with >50% mucinous areas (14%) had significantly poorer prognosis than non‐mucinous in stages A and C, while mucinous differentiation did not correlate with prognosis in stages B and D. Lymph nodes regional to mucinous tumours had significantly less paracortical response, and those with >50% mucinous differentiation, significantly less perivascular lymphocyte cuffing at the tumour margins. These lymph node and stromal compartments are putative T‐lymphocyte areas, and hence our findings suggest that mucinous tumours are either less stimulatory or perhaps inhibitory of cell‐mediated immunity.

Palliative operative management in rectal carcinoma

The results of palliative operative management of 338 patients with rectal carcinoma managed by one of the authors are presented. Postoperative mortality was higher for patients undergoing palliative resection (11.7 per cent) than colostomy bypass (5.3 per cent) or diagnostic laparatomy (6.8 per cent). Cancer specific survival following palliative resection was significantly (P<0.001) longer than that following colostomy bypass or diagnostic laparotomy for tumor Stages D1 (local visceral involvement) and D2 (distant metastases). However, in patients with liver or peritoneal metastases alone, cancer specific survival did not differ significantly after the operations of resection or colostomy bypass. The failure to demonstrate improved survival after resection of the primary tumor in these latter two groups with distant metastases indicates the dominant role of volume of tumor tissue present in these situations. The results suggest that longer survival following pallitative resection reflects a bias of patient selection towards more favorable cases.

The risk of rectal carcinoma following colectomy in ulcerative colitis.

In a series of 1439 patients with ulcerative colitis, managed by one of the authors (E.S.R.H.), surgical resection was performed on 374 patients (26 per cent): colectomy, 273 (subtotal colectomy and mucous fistula, 172, colectomy and primary ileorectal anastomosis, 101); proctocolectomy, 61; and miscellaneous procedures, 40. Of the 172 patients undergoing subtotal colectomy and mucous fistula, 93 (54 per cent) subsequently required rectal excision, 33 (19 per cent) had ileorectal anastomosis performed as a second procedure, and in 46 (27 per cent) subsequently developed as a mucous fistula. Two hundred seventy-three patients were at risk for the development of rectal, cancer after subtotal colectomy; ten patients (3.6 per cent) subsequently developed rectal cancer. The cumulative probability of developing rectal cancer after subtotal colectomy reached 17 per cent at 27 years from disease onset. The tumors were more advanced in stage and of higher grade malignancy than those of a parallel general series of patients with rectal cancer uncomplicated by inflammatory bowel disease. Colectomy and ileorectal anastomosis has been successful for most patients. However, the experience of this series highlights the danger of carcinomatous transformation in the retained rectum, the requirement for regular long-term follow-up, the need for markers of precancerous change, and the value, where relevant, of prophylactic proctectomy.

Adenocarcinoma of an ileostomy in a patient with ulcerative colitis

A case is reported of adenocarcinoma of the ileum following surgical management of ulcerative colitis. In this patient it seems likely that the carcinoma arose in a pre-existing tubulovillous adenoma of rectal origin. This case report draws attention to the need to exclude any possibility of retention of large-intestinal mucosa in the ileostomy when transecting the ileum, either at primary colectomy or when dismantling an ileorectal anastomosis.

Immunohistological patterns of carcinoembryonic antigen in colorectal carcinoma. Correlation with staging and blood levels

&NA; Forty‐four primary adenocarcinomas of the large bowel and 2 liver metastases were stained for carcinoembryonic antigen (CEA) in tissue sections by indirect immunofluorescence. All tumours were positive and showed either one or more of 3 different patterns‐luminal; linear at surface of the tumour cells; cytoplasmic. In most cases (83%), two or all 3 patterns were seen in the same or in different parts of a tumour. The immunohistological staining was concordant with preoperative blood levels of CEA in 31 cases (67%) in that 26 tumours showed strong immunofluorescence associated with blood CEA above 2.5 μg/l, and 5 showed weak staining and blood CEA values less than 2.5/μg/l. However, in 7 strong staining was associated with low blood CEA, and in 8 weak staining was associated with high blood levels. The dissociation between histological and blood CEA findings in 1/3 of the cases, together with the marked variation within the same tumour and differences between one of the primaries and its liver recurrence, suggest that CEA immunohistology is of no better prognostic value than blood CEA levels. There was no association between CEA immunohistology and tumour staging or differentiation. However, blood CEA levels were significantly higher in tumours with extensive local or distant spread (stage D) and in poorly differentiated tumours.

T-Antigen Expression by Peanut Agglutinin Staining Relates to Mucosal Dysplasia in Ulcerative Colitis

Staining of 326 rectal mucosal biopsies from ulcerative colitis patients with peanut agglutinin (PNA), which binds to the T-blood group antigen and has been claimed to reflect a cancer-associated mucin alteration, showed highly significant direct associations with mucosal dysplasia (P<0.001), disease activity (P<0.001), and subsequent development of rectal cancer in a smaller series of patients (P=0.005). Staining for normal colonic mucin by theDolichos biflorus (DBA) lectin related significantly and inversely to dysplasia. Intense normal colon mucin staining by DBA related significantly (P<0.025) to long disease duration and to subsequent development of cancer (P=0.02). The latter association is based on a small number of patients only and is not considered conclusive evidence, but may provide a link with goblet-cell hyperplasia. The authors conclude that although T-antigen expression relates to dysplasia, the findings of “false” positive and negative rates of 22 and 33 percent respectively, make it unlikely that staining of biopsy sections for the T-antigen by peanut agglutinin will contribute materially to routine assessment for dysplasia and cancer risk prediction in patients with ulcerative colitis.

Resected ovarian recurrence from colorectal adenocarcinoma: A study of 13 cases

Thirteen (1.4 per cent) of 882 female patients managed by resection for primary colorectal adenocarcinoma subsequently required operation for ovarian recurrence. The clinical, pathologic and survival data of this group have been analyzed. Their age (mean±s.d., 51.2±8.4 years) was less (P=0.004) than that of the total series of female patients (59.4±13.0 years). Survival varied from 15 to 96 months (mean 17 months) from the ovarian operation. Three patients were still alive with no evidence of recurrence at the time of review. The low incidence of clinical ovarian recurrence requiring operation does not support recommendations for prophylactic oophorectomy in all patients.

A major predictor of cancer following ileorectal anastomosis

Fifty patients with ulcerative colitis managed by colectomy and ileorectal anastomosis had rectal biopsies performed in the period 1967 to 1972. Follow-up information was available on all patients. Thirty-nine patients were reviewed and rectal biopsies performed in the 1980 to 1982 period. Three patients had developed rectal cancer in the period 1975 to 1980, and two rectal cancers were detected in the 1980 to 1982 follow-up period. All cancers occurred in patients with a diagnosis of moderate or severe dysplasia in biopsy specimens from the 1969 to 1972 period. The probability of developing rectal cancer after a diagnosis of moderate or severe dysplasia in this series reached 42 per cent at nine years from diagnosis.