Prognostic factors of colorectal cancer
With regard to survival, many studies have been conducted on various clinical, biological, histological, and molecular factors in the hope of identifying useful indicators of clinical outcome to correctly estimate its prognosis and thus help in the formulation and implementation of comprehensive treatment plans. Although many factors are associated with prognosis, the results are not uniform, and the Dukes and TNM staging norms proposed in 1932 are still the most desirable prognostic indicators for colorectal cancer. No other clinical tumor, such as colorectal cancer, has such a close correlation between prognosis and pathological diagnosis and staging.
I. Clinical factors affecting prognosis
(I) Gender
Among patients with colorectal tumors, the prognosis varies between male and female due to anatomical and physiological differences, and most reported that female patients have a better prognosis than male patients, especially in rectal cancer [47, 48]. The reasons may be related to the higher incidence of men than women, longer natural survival of women than men, and the protective effect of estrogen [49]; in rectal cancer, the difference in pelvic anatomy and physiology makes radical surgery easier for women than for men.
In 3751 cases of colorectal cancer in the Cancer Hospital of Sun Yat-sen University, the composition ratio of female patients increased from 38.7% to 42.6% from the 1960s to the 1990s, and the 5-year survival rate also tended to decrease compared to male patients [1].
The Institute of Oncology, Zhejiang Medical University analyzed 743 cases of colorectal cancer and the overall 5-year survival rate, male (58%) and female (60%) did not differ significantly. However, when combined with age and Dukes’ staging, there was a significant difference in both [50].
(ii) Age
The incidence of colorectal cancer in China is lower than that in Europe and the United States, but the median age is about 55 years, which is 12-18 years earlier than that in Europe and the United States, and colorectal cancer in young people is relatively more frequent and the prognosis is worse than that in the middle and old age groups. The 5-year survival rate after surgery was reported to be 40.1% in young people and 51.4% in elderly patients in China [51]; the 5-year survival rate was 21.8% in the young group and 53.0% in the middle-aged and elderly group in Shanghai Cancer Hospital; the 5-year survival rate was 31.9% in those younger than 30 years old and 63.8% in the group older than 65 years old in Cancer Research Institute of Zhejiang Medical University [50].
It is generally believed that the main reasons for the poor prognosis of colorectal cancer in young people may be related to late diagnosis, poor tumor differentiation, more indolent cell carcinoma, rapid tumor growth and early lymph node metastasis [48, 50, 52-55].
The effect of age on the survival rate of colorectal cancer patients has also been reported to be uncertain and controversial at home and abroad. a study by Cerottini [56] concluded that patients with intestinal cancer aged <50 or >70 years had a poor prognosis and that age was an independent prognostic factor affecting the prognosis of colorectal cancer. Wang et al [57] analyzed 151 patients and concluded that there was no difference in pathological staging and prognosis between the young and old groups and that age was not a determinant of colorectal cancer prognosis.
The differences in conclusions may be related to the definition of subgroups of young people, ethnicity, cultural environment and the prevalence of health education. This draws our attention to the need to enhance the prevalence of health education in the prevention of the disease.
(iii) Disease duration and clinical symptoms
In general, the longer the duration of symptoms and the later the consultation time of tumor patients, the worse their prognosis [58]. However, it is also believed that patients with symptoms lasting longer than 6 months have a significantly higher rate of radical surgery, lower postoperative mortality and higher 5-year survival rate because the tumors in these patients have a slower growth rate, lower malignancy and a relatively better prognosis, while those with a short duration of disease and obvious symptoms have a more rapid tumor development, poor pathological differentiation and high malignancy.
(iv) Complications
The prognosis of patients with complications is poor, and there are differences in the prognosis of different complications.
1. Bleeding
Tumor secondary infection can appear mucus and blood stool, while tumor bleeding can be seen as jam-like stool or bloody stool or fresh blood stool. The prognosis of patients with bleeding as the main clinical symptom is better than those with bleeding as the main clinical symptom, as reported by Thoms [61].
2. Intestinal obstruction and perforation
Patients with combined obstruction and perforation indicate that the tumor has developed to the middle and late stage, which not only brings difficulties in surgery and treatment, but also increases the surgical mortality rate and its prognosis is poor. Postoperative in-hospital mortality rates of 39% and 53%, respectively, have been reported in those with such complications [62]. NSABP supports a strong prognostic impact of intestinal obstruction. analysis by GITSG [63] found that obstruction was an important factor affecting prognosis independently of Dukes’ stage. For disease-free survival intestinal perforation is as important as obstruction.
(V) Characteristics of the primary tumor
1.Site of primary tumor
The site of colorectal cancer is related to the prognosis, and the prognosis of rectal cancer is worse than that of colon cancer, and the prognosis of right hemicolectomized colon cancer is the best [64]. It is also believed that tumor site is one of the independent prognostic factors of colorectal cancer [65], but there are different views [66, 67]. The 5-year survival rate after surgery for colon cancer is much higher than that for rectal cancer, 77% and 66%, respectively, according to Zhongshan University Cancer Hospital. This may be related to the anatomical characteristics of rectal cancer, which is more prone to peri-intestinal infiltration and metastasis and not easily resected completely, etc. The results of Pihl et al. on the 5-year tumor-free survival rate of each Dukes stage of colon and rectal cancer also support that the main difference between the survival rates of the two sites lies in stage C.
As far as rectal cancer is concerned, tumors located below the retroflexion, without peritoneal envelope, are easy to invade and have less chance of radical treatment. If they invade the pelvis, vagina or prostate, the prognosis is obviously poor; its lymph node metastasis pathway not only flows back with mesenteric lymph, but also can metastasize to both sides via internal iliac artery parietal and sciatic rectal fossa lymph nodes; and the pelvic floor is rich in vascular network, so the possibility of hematologic metastasis is more obvious than that above the retroflexion, so the prognosis is poor.
2.The size of primary tumor
It is generally believed that the size is directly related to the depth of tumor infiltration and metastasis [47, 68, 69]. According to the data of the National Collaborative Group, the 5-year survival rate of tumors less than 2 cm is as high as 73.2%, while that of those larger than 5 cm is only about 50%.
3.Shape and growth pattern of primary tumor
The shape and growth mode of colorectal cancer can reflect the biological characteristics of the tumor to some extent, the polyp type or exophytic type that bulges into the intestinal cavity is less likely to invade and penetrate deep into the intestinal wall and less likely to have lymphatic and hematogenous metastasis, while the broad-based infiltrative growth and ulcerated tumor are very likely to have lymphatic and hematogenous metastasis. The GITSG adjuvant colon study compared exophytic and non-exophytic tumors and found that exophytic lesions had a very significant positive effect on survival [63]. Domestically, they were classified as cauliflower, ulcerated and infiltrated types, and their 10-year survival rates were 51%, 39% and 21%, respectively, with statistically significant differences among the three types.
(vi) Blood transfusion
The relationship between blood transfusion during surgery and increased recurrence of colorectal cancer is unclear, and it is generally believed that transfusion increases postoperative recurrence of metastasis by decreasing immunity and increasing postoperative infectious complications [68,71].A multifactorial analysis of 1051 patients with colorectal cancer at Mayo Hospital showed that blood transfusion during surgery did not decrease disease-free survival, but it decreased overall survival [72].
Busch et al [73] reported that patients were randomized prior to surgery to receive an intraoperative transfusion of autologous or allogeneic blood (if needed), and there was no difference in disease-free survival between the two groups. Subgroup analysis showed that patients who had donated blood but had never received a transfusion had higher disease-free or cancer-related survival than those who had received a transfusion. This suggests that the factors that make transfusion a must are perhaps more important than the transfusion itself.
II. Pathological factors
(i) Histologic type and degree of differentiation
The histological type reflects the biological nature of the tumor and is the most basic factor in determining the prognosis, which is closely related to the degree of differentiation, and the grading method proposed by Grinnell based on the arrangement of tissue glands, nuclear polarity and nuclear division is the most widely used. Generally, the tissue types are categorized into three major groups: highly differentiated group (highly differentiated adenocarcinoma, papillary adenocarcinoma), which has the best prognosis; moderately differentiated group (moderately differentiated adenocarcinoma, mucinous adenocarcinoma), which has the second best prognosis; and poorly differentiated group (poorly differentiated adenocarcinoma, indolent cell carcinoma, undifferentiated carcinoma), which has the worst prognosis. Among them, the worst prognosis is in the indolent cell carcinoma and the best prognosis is in the carcinoid [74, 75].
(II) Lymph node metastasis
The presence or absence of lymph node metastases and the number of lymph node metastases have a great influence on the prognosis [1] (Table 16-3-3). Patients without lymph node metastases are mostly cured with a 5-year survival rate of about 70%, while those with lymph node metastases are reduced to less than 30%. The number of lymph node metastases is negatively correlated with the prognosis, and the higher the number, the worse the prognosis. However, the positive detection rate of lymph nodes is related to the scope of surgical resection and whether the lymph nodes are examined carefully; if conditions permit, it is better to examine after removing all the fat.
(C) Depth of infiltration
The depth of tumor infiltration is closely related to the prognosis, the deeper the infiltration, the worse the prognosis. The 5-year survival rate of tumor infiltration not reaching the muscular layer can reach more than 90%, and those who do not break through the muscular layer also have 70%, but those who infiltrate the plasma membrane or extra-plasma membrane infiltration drop to 30%-40%. It may be related to the richness of lymphatic vessels and blood vessels in the plasma membrane layer, once the tumor penetrates into the subplasma layer, it is very easy to accelerate the spread of lymphatic and blood channels.
(iv) Vascular infiltration
There are many differences regarding the definition of vascular infiltration, methods of detection, and metastatic potential of cancer cells that invade blood vessels or lymphatic vessels, which cause differences in the observed results.
1.Vascular infiltration (BVI)
The prognosis is worse when the tumor tissue infiltrates the blood vessels. Among them, veins are most likely to be involved, while arteries are rarely invaded, and the overall incidence does not exceed 1%.
(1) In terms of site, there are: 1) vascular invasion within the intestinal wall called intramural BVI; 2) vascular invasion outside the intestinal wall (peri-colonic adipose tissue and epithelium) called extramural BVI. usually BVI mainly refers to venous infiltration. the incidence of BVI ranges from 25% to 81% and increases with stage and grade [76-78]. chapuis [76] et al. considered BVI as an independent prognostic factor. Krasna [79] and Inoue [77] et al. applied elastic tissue staining and scoring to identify BVI and lymphovascular infiltration (LVI) in two groups of cases.Krasna reported a significantly lower 3-year survival rate in BVI-positive patients than in negative patients (30% : 62%), a significantly higher rate of extramural lymph node metastasis in BVI-positive patients than in negative patients ( Inoue et al. found that the incidence of BVI was higher in patients who died within two years (61% : 31%).
(2) In terms of infiltration mode, tumor infiltration of blood vessels can be: (1) cancer thrombus; (2) cancer cells adhering to vascular epithelium; (3) cancer tissue destroying vascular wall. The recurrence rate, metastasis rate and survival of patients with and without vascular infiltration were also significantly different even if the Dukes stage was the same.
The incidence of LVI ranged from 8% to 73% and increased with stage and grade. increased incidence of LVI was associated with decreased survival, and LVI was an independent prognostic factor
Michelassi et al [81] examined 110 patients with rectal cancer and found that 73% had microscopic infiltration of blood vessels or lymphatic vessels and an increased rate of local recurrence in LVI-positive patients (23% : 0%).
Minsky et al [82] analyzed 462 patients with colorectal cancer and found an increased incidence of LVI in colonic tumors compared to rectosigmoid and rectal tumors (15% : 10%). Patients with extramural and intramural BVI had the highest incidence of LVI compared to BVI-negative tumors (52% : 5%). the relative number of lymph node metastases was significantly higher in LVI-positive patients than in negative patients (59% : 25%). The 5-year survival rate was significantly lower in LVI-positive patients than in negative patients (57% : 84% in colon; 38% : 71% in rectum) for both colon and rectal tumor patients.
(V) Degree of mesenteric spread and peripheral (radial) cut margin
The staging of rectal cancer is based on whether tumor infiltration penetrates the bowel wall, which logically means examining the margins on both sides (radial) of the resected specimen. The degree of mesenteric spread of rectal cancer is closely related to the peripheral (radial) margins, which is an independent prognostic factor affecting survival.
And it was found that peripheral margin involvement was found in 20-33% of patients with proximal and distal margin-negative rectal cancer [83, 84].Preliminary results of the Dutch study suggest that postoperative radiotherapy also does not reduce the recurrence rate of margin-positive rectal cancer. Peripheral margins may be the single most important clinicopathological factor determining the risk of local recurrence in rectal cancer compared to distal margins.
(vi) Interstitial tumor reaction
As early as 1934, the relationship between tumor interstitial and marginal lymphocyte reaction and prognosis was noticed. (1) Most studies have concluded that it is a positive prognostic factor regardless of univariate or multifactorial analyses [85-87]; (2) Jass et al [85] confirmed that lymphocytic infiltration is not only the most important factor in grading models, but also has an important role in “optimal” models where it is used as a relevant parameter for grading and staging. The role of lymphocyte infiltration is not only the most important factor in grading models, but also in the “best” models where it is used as a relevant parameter for grading and staging.
In China, the lymphocytic reaction in the interstitial and marginal tumors is classified into 3 degrees: degree 0 is no or only individual lymph node infiltration; degree I is a significant lymphocytic reaction, ranging from small to moderate amounts; degree II is the presence of a large number of lymphocytes, and even lymphoid follicles can be formed. The 2-year, 5-year and 10-year survival rates differed significantly among the three.
(vii) Perineural infiltration
The incidence of perineural infiltration ranges from 14% to 32% and is associated with vascular invasion. seefeld and Bargen pointed out that tumor infiltration along the perineural space can spread to 10 cm from the primary tumor. the incidence of perineural infiltration is 30% and increases with grade and Dukes stage. Most studies consider nerve infiltration as an independent prognostic factor for survival [88-90].
(viii) Nuclear morphology
Nuclear morphometry has been used to determine whether survival in patients with colorectal cancer is associated with differences in nuclear size and shape, with variable results. Mitmaker et al. applied multiple regression analysis and found that nuclear shape was the most effective prognostic indicator for survival in colorectal cancer patients. In contrast, Heimam et al. reported no correlation between nuclear morphology and staging or survival. ambros analysis showed that although there was no association between nuclear intermediate zone and survival, it showed a significant linear correlation with the number of positive lymph nodes in colorectal cancer.
(ix) Distant metastasis
Colorectal cancer has a large proportion of distant metastasis, and when distant metastasis occurs, its prognosis is poor. The most common site of metastasis is liver. 10%-25% of patients have liver metastasis at the time of diagnosis, and about 16%-22% of patients have liver metastasis after radical surgery. Next is the lung, and liver metastases have a worse prognosis than lung metastases. If liver metastases are isolated lesions, the survival rates at 1, 3 and 5 years after surgical resection are 80%, 55% and 36%, respectively; among them, the 5-year survival rates after resection of simultaneous and heterochronic liver metastases are 27% and 31%, respectively. Other metastases such as bone, brain, and left supraclavicular lymph nodes are usually incurable if they occur. Brain metastases are not uncommon, about 10%, and the prognosis is even worse.
(X) Staging
After radical surgery for colorectal cancer, the independent pathological factor that is crucial to survival and recurrence rate is tumor stage. The clinicopathological staging of colorectal cancer integrates the depth of infiltration of the intestinal wall, the degree of lymph node involvement and the presence of distant metastasis, reflecting the biological behavior of the tumor and its progression. The purpose of clinicopathological staging is to provide reference for choosing treatment plan and to accurately estimate the prognosis. Commonly used are Dukes staging and the internationally accepted TNM tumor staging, which is currently considered the closest reference for evaluating the prognosis of colorectal cancer [1] (Table 16-3-5).
It is officially recommended by BP and other important clinical institutions.
In conclusion, the factors affecting the prognosis of colorectal cancer are comprehensive. The type of tissue and the degree of differentiation can affect the biological behavior of the tumor, including the basic factors such as how the tumor grows, how fast it grows, and whether it is prone to lymph node metastasis, but specifically for a patient, then the clinicopathological stage of the tumor at the time of consultation is the main basis for determining the prognosis.