Navigation

Esophageal Cancer Survivors

From January 1969 to December 1980, 8,948 patients with esophageal carcinoma were registered in Japan. Among these patients, 5,506 underwent resection. The ten-year survival rate was 12.3 percent for all registered patients and 18.7 percent for resected cases. Female patients had significantly better survival rates than male patients. Depth of invasion correlated better with ten-year survival than the superficial extent of the tumor. The TNM classification revised in 1987 was examined in relation to the survival, and it was found to have good prognostic value.

Carcinoma of the esophagus is one of the most difficult carcinomas to cure, even in supposedly curative resectable cases. Although preoperative and postoperative radiotherapy or chemotherapy also have been performed during the past 20 years in an attempt to improve therapeutic results, the prognosis is still poorer than that for other carcinomas.

There are few papers reporting long-term survival after treatment. This article describes the ten-year survival rate of patients with esophageal carcinoma registered in Japan between 1969 and 1980.

MATERIALS AND METHODS

In Japan, since the Guidelines for Clinical and Pathologic Studies on Carcinoma of the Esophagus[1] was published in 1969, findings of patients with esophageal carcinoma have been described according to these guidelines. This has simplified collection of data on patients from many institutions in Japan. From January 1969 to December 1980, 10,113 patients with esophageal carcinoma were registered at 234 institutions. To investigate the survival rate, long, accurate follow-up of patients is essential. Seventy-eight institutions were excluded because their five-year follow-up rate was under 80 percent. Thus, 8,948 patients were included in this study; among these, 5,481 patients underwent resection. Out of 5,481, 132 patients had cancer of the esophagogastric junction and 82 with cancer of the cardia had adenocarcinoma; the remaining 5,267 patients with cancer of the esophagus had squamous cell carcinoma. The survival rates of these patients were computed according to the life-table method of Cutler[2] with the use of a HITAC M 160-H (Hitachi Ltd., Tokyo). The significant difference between each survival rate was computed to be in the range of 95 percent confidence limits by Greenwood's formula.[2]

RESULTS

The survival rate of the registered patients is shown in Table 1. The survival rate was 15.3 percent at five years and 12.3 percent at ten years. While there were more male patients, with the male-female ratio being about 5:1, female patients had better survival rates than males for each year of survival, and the difference was statistically significant. [TABULAR DATA OMITTED]

The survival rate of 5,481 resected cases was 23.8 percent at five years and 18.7 percent at ten years (Table 2). Female patients had significantly better survival than males at five and ten years. [TABULAR DATA OMITTED[

Table 3 shows the correlation between the location of the tumor and survival. Cases with lesions in the cervical, lower thoracic and abdominal esophagus had a better survival rate than cases with lesions in the upper thoracic esophagus. Patients with middle esophageal carcinoma had intermediate survival. While these results agreed with other reports, there were no statistically significant differences among the survival rates for lesions at different sites. [TABULAR DATA OMITTED]

The relationship between the length of the tumor and survival is shown in Table 4. Tumors less than 1 cm showed the best survival, followed by those less than 3 cm. When the length of tumor was greater than 3 cm, the survival rate decreased, but there was no statistically significant difference in tumors within the 3- to 10-cm range. Among cases of tumors more than 15 cm, there was no survivor beyond four years. [TABULAR DATA OMITTED]

Among resected cases, operated specimens were examined histopathologically to evaluate the depth of invasion and lymph node metastasis.

The depth of invasion was classified as being as far as the intraepithelium, muscularis mucosa, submucosa, muscularis propria, invasion reaching the adventitia, definite invasion to the adventitia, invasion to neighboring structures according to the Japanese guidelines (Table 5). There was a decrease in survival with the depth of invasion and significant differences were observed among tumors whose depth of invasion reached the intraepithelium, muscularis mucosa and submucosa and submucosa and muscularis propria. Clearly the worst survival rate was observed in cases of tumors that had invasion to neighboring structures. Resected cases, which accounted for 60 percent of registered cases (Table 6) had apparently better survival than nonresected cases. [TABULAR DATA OMITTED]

Operations were classified as curative or noncurative resection according to the Japanese guidelines.[1] Curative resection means complete removal of the primary tumor and dissection of regional lymph nodes. Other operations were classified as noncurative. Both groups were divided into two subgroups, based on whether the patients received reconstructive surgery or not. The reason why reconstruction was not possible in some cases was mainly due to low pulmonary function and pneumonia. Resection and reconstruction were performed in one stage in the majority of cases. There was a clear difference of survival between the curative and noncurative groups. In curatively resected patients, if they did not receive reconstruction, survival decreased to 15.6 percent at five years and 10.7 percent at ten years, compared with 32.7 percent at five years and 26.3 percent at ten years in those receiving reconstruction surgery (Table 7). There was no difference in operative mortality (within 30 days) between these groups. [TABULAR DATA OMITTED]

Table 8 shows the results of radiotherapy alone. Radical radiotherapy means that the planned dose (50 Gy) was given to the patients. Because of the selection for treatment, survival in the radiotherapy-only group is low, even in the radically treated group. The survival of the radical radiotherapy group was 5.3 percent at five years and 3.0 percent at ten years. Table 8--Radiation Therapy Alone

                                    Survival Rate (%)
               No.
             Patients   1 yr   2 yr   3 yr   4 yr   5 yr   10 yr
Total         1,532     29.5   12.2   7.1    5.3    4.5    2.5
Radical         911     37.1   16.3   9.1    6.7    5.3    3.0
Palliative      621     18.3    6.0   4.1    3.3    3.2    1.0

In the resected cases, the extent of disease was examined based on the 1987 TNM classification.[3] Table 9 shows the relationship between tumor and survival. The T1 classification had the best survival followed by T2, T3 and T4. There were statistical differences in survival for each category. [TABULAR DATA OMITTED]

Table 10 indicates the survival according to lymph node metastasis. There was a statistically significant difference in survival between N0 and N1. [TABULAR DATA OMITTED]

Table 11 shows survival and metastasis classification. The M1 classification includes not only organ metastasis but also metastasis to nodes more distant than regional nodes. There was a clear difference between M0 and M1, but there were some patients surviving more than ten years in the M1 group. [TABULAR DATA OMITTED]

Stage I had the best survival rate: 64.2 percent at five years and 48.0 percent at ten years (Table 12). There was a clear difference between each of the five groups, suggesting that this staging reflected survival well. [TABULAR DATA OMITTED]

Radiation therapy, including preoperative irradiation, has been used in combination with surgery. There was no difference in the five- and ten-year survival rates of the nonirradiated and preoperative radiation groups. The postoperative radiation group had a slightly lower survival than the other two groups (Table 13). [TABULAR DATA OMITTED]

The histologic effect of radiation recognized on the resected specimens and survival is shown in Table 14. Histologic effect was classified as follows according to the Japanese guidelines.[1] Markedly effective: cancer is eliminated and no viable cancer cells are observed. Moderately effective: viable cancer cells occupy less than one third of the lesion, with destructive cells in the rest of the lesion. Ineffective or slightly effective: viable cancer cells occupying more than one third of the lesion. The markedly effective group had the best survival (31.0 percent) at ten years. On the contrary, the slight effect group had the worst survival, with a ten-year survival of 12.8 percent.

The causes of death after surgery are shown in Table 15. In the first two years, the main cause of death was due to the primary disease, esophageal carcinoma. However, after five years death unrelated to malignancy represented about two thirds of those who died.

DISCUSSION

There are only a few reports concerning long-term survival of patients with cancer of the esophagus. Earlam and Cunha-Melo[4] collected 83,783 patients from the literature, and reported a 12 percent five-year survival for resected cases, and an 18 percent survival for those leaving the hospital after resection. Kinoshita et al[5] evaluated ten-year survival after resective surgery and reported 58 ten-year survivors among 1,329 radically resected cases. K'ai and Huang[6] reported a 21.2 percent ten-year survival rate among 652 resected cases. There is no other report on ten-year survival. In Japan, multi-institute registration of patients with esophageal carcinoma treated after 1969 was commenced in 1975. In 1985, we proposed a new TNM classification for esophageal carcinoma based on 3,681 resected cases registered from 1969 to 1978.[7] We stressed that the depth of invasion had a better correlation with five-year survival than length or circumference of the tumor. We also emphasized that perigastric lymph nodes should be included in the N category rather than the M category, based on the correlation with the five-year survival rate. [TABULAR DATA OMITTED]

The present study included 10,113 registered cases, 5,506 of which were resected. The survival rate was calculated by the life table method for up to ten years. The overall survival of resected cases was 23.8 percent at five years and 18.7 percent at ten years. These results are better than those of previous reports in Japan.

Female patients had better survival than males for each year of survival, both overall among all registered cases and also among resected cases.

Concerning the location of the tumor in the resected cases, those with cancer of the upper thoracic esophagus had the worse survival, but there was no marked difference between those with lesions in the middle or lower thoracic esophagus.

As to the length of the tumor, cases with tumors less than 3 cm had better survival, but there was no difference among those in the 3- to 10-cm category. Thus, in the present series the length of 5 cm had no significance as a benchmark for survival.

Depth of invasion clearly reflected survival. If the carcinoma invasion was limited to the muscularis mucosa, the ten-year survival rate was 50 percent. However, if invasion reached the submucosa, the ten-year survival dropped to 36.4 percent, which was not very different from musularis propria (25 percent) or invasion reaching the adventitia (22.5 percent) cases. This finding suggests that submucosa invasion is an indicator of advanced stage in esophageal carcinoma, and that treatment should be performed accordingly.

Comparing patients with or without resection, the former had better survival. This indicates that patients with resectable tumor should undergo surgery.

Among the resected cases, patients undergoing reconstruction had better survival than those without reconstruction in both curative and noncurative resection categories. There were long-term survivors among the patients who did not receive reconstructive surgery, and it is necessary to assess the patient's status in terms of whether reconstructive surgery is possible or not, because this is closely related to the patient's quality of life after surgery. [TABULAR DATA OMITTED]

In patients receiving resection surgery, the revised TNM classification[3] was examined in terms of correlation with survival. The tumor classification, based on the depth of invasion, clearly reflected the survival, as seen in Table 9. The T1 cases had 49.9 percent survival at five years and 36.1 percent at ten years. There is a clear difference of survival between each tumor group. There is also a statistically significant difference between lymph node classifications N0 and N1. Because patients with metastases to more distant lymph nodes had definitely worse survival, there should be no objection to this being included in M1.

In the metastasis category, there were some patients with M1 classification who survived ten years. There remain some problems concerning the definition of M1 because M1 includes not only organ metastasis but also distant lymph node metastasis.

Furthermore, staging based on the TNM classification correlated well with survival. The stage 1 group had 64.2 percent survival at five years and 48.0 percent at ten years. As the stage advanced from 1 to 4, the survival decreased with statistically significant differences.

Radiation frequently is combined with surgery, including preoperative irradiation, but the survival rate was lower than for the nonirradiated group. This may be due to patient selection before treatment. Furthermore, the postoperative irradiation group had lower survival than the preoperative irradiation group. This result reflects the fact that mostly patients with lymph node metastasis received postoperative radiotherapy. To clarify this, we performed a cooperative randomized trial comparing preoperative and postoperative irradiation. In this trial, the postoperative radiation group had better survival than the preoperative radiation group.[8]

In the preoperatively irradiated group of cases, those in which marked effect was observed on the resected specimen had better survival than those with only slight effect. Locally effective radiation therapy enhanced the survival after surgery.

As seen in Table 8, the radiation therapy-alone group had worse survival than the surgery group, probably due to selection before treatment, since more patients in the advanced stage were included in this group. Randomized trials comparing surgery and radiation are necessary.

REFERENCES

[1] Japanese Society for Esophageal Disease. Guidelines for clinical and pathologic studies on carcinoma of the esophagus. Jap J Surg 1976; 6:69-86

[2] Cutler SJ, Ederer F. Maximum utilization of the life table method in analyzing survival. J Chron Dis 1958; 8:699-712

[3] International Union Against Cancer. TNM classification of malignant tumors. 4th fully revised ed. Berlin: Springer-Verlag, 1987

[4] Earlam R, Cunha-Melo JR. Oesophagus squamous cell carcinoma, a critical review of surgery. Br Surg 1980; 64:457-61

[5] Kinoshita Y, Nakayama K, Endo M, Sato H, et al. Evaluation of ten year survival after operation for upper and mid-thoracic esophageal cancer. Int Adv Surg Oncol 1978; 1:173-200

[6] K'ai WY, Huang GJ. Surgical treatment. In: Huang GJ, K'ai WY, eds. Carcinoma of the esophagus and gastric cardia. Berlin: Springer Verlag, 1984:275-84

[7] Iizuka T, Akiyama H, Isono K, Endo M, Kakegawa T, Mori S, et al. A proposal for a new TNM classification for carcinoma of the esophagus. Jap J Clin Oncol 1985; 14:625-36

[8] Iizuka T, Ide H, Kakegawa T, Sasake K, Takagi I, Ando N, et al. Preoperative radioactive therapy for esophageal carcinoma, randomized evaluation trial in eight institutions. Chest 1988; 93:1054-58

COPYRIGHT 1989 American College of Chest Physicians
COPYRIGHT 2004 Gale Group