| AimsIt was reported by foreign researchers that patients with hereditary predisposition makes up approximately 20 percent of all patients with CRC. With regard to hereditary predisposition of CRC in CRC patients in China, there were some reports on patients from the southern district. But till now, the condition of hereditary predisposition of malignant tumors (especially CRC) on the northern patients hasn't been reported. Hereditary nonpolyposis colorectal cancer (HNPCC) is the most common form of hereditary CRC in an autosomal dominant manner. Most of reports on clinical phenotype of HNPCC were case reports, and the studies on prevalence of HNPCC were only restricted in some areas. Moreover these studies have many problems such as inconsistent criteria, small samples and a small area. It is difficult to get a systematic understanding on clinical phenotype of HNPCC and prevalence of HNPCC in CRC patients, and to define a set of criteria which are more favorable to the situation of our country. The unit of HNPCC syndrome is a family, but not a patient. So we should discuss the phenotype of this syndrome according to family unit. All conclusions of HNPCC phenotype are currently drawn by analyzing patients with HNPCC-associated tumor, which can only provide general rules of HNPCC phenotype. But they couldn't provide more information in order to deeply study genetic pathogenesis of HNPCC syndrome and make a correlation between its phenotype and its genotype. It hasn't been reported that the description of HNPCC phenotype is based on family units till now.It is generally accepted that dysfunction of DNA mismatch repair system is correlated with genetic pathogenesis of HNPCC syndrome, which is resulted from germline mutation of mismatch repair genes. It is important to detectgermline mutation of mismatch repair genes in typical and suspected HNPCC families. It can not only help us to explore the role of germline mismatch repair genes mutation in genetic pathogenesis of HNPCC syndrome and reveal the new pathogenesis of tumors, but also make genetic test to help diagnosis and prevention of HNPCC syndrome. The studies from foreign countries showed detecting germline mismatch repair genes mutation (mainly hMLHland hMSH2) had an important role in predicting cancer risk of members (especially ones with first-degree relatives with malignant tumors ) in HNPCC families and early detecting and early treatment of HNPCC-associated tumors, especially CRC and cancer of endometrium. But up to now, it hasn't been reported that cancer risk of members in HNPCC families is predicted by germline mutation of mismatch repair in China. According to the range of mutation, mismatch repair genes mutation can be classified as two groups: minor mutations and large genomic deletions/insertions. The majority of mutations were minor mutations, which were found in typical and suspected HNPCC families. And minor mutations were 70% or so in all found mutations of mismatch repair genes. Currently, there are many methods which are used in the study of mismatch repair mutations. But they all couldn't detect minor mutation and large genomic deletions/insertions at the same time. PCR-SSCP, DGGE and DNA sequencing are the methods used in detecting minor mutations. But both PCR-SSCP and DGGE are low sensitive, and DNA sequencing is time-consuming, laborious and costly. They aren't clinically used in screening mutation on a large scale. DHPLC is a new technology of screening DNA variation which has many merits such as automatic, quick, high-effective, detecting a larger fraction of DNA mutation. DHPLC can detect replacement, insertion and deletion of single nucleotide. And its sensitivity and specificity are both more than 96%. Although the results should be confirmed by DNA sequencing, it can significantly reduce the cost. The sensitivity and specificity of DHPLC have a dependence of genes in detecting DNA mutation. DHPLC often fail to the mutation located in high melting domains embedded in low melting DNA regions. When DHPLC detected mutations of hMLHl and hMSH2, the sensitivity was more than 97% and the specificity varied greatly in the reports from other countries. Mutations ofmismatch repair genes have recently been detected by DHPLC in China. But DHPLC was directly applied to detect the mutations of mismatch repair gene and its sensitivity and specificity were never evaluated in our country. It hasn't been reported that the sensitivity and specificity of screening the mutations of hMLHl and hMSH2 was evaluated by the results of DNA sequencing. So up to now, its correctness and efficiency haven't been known in screening the mutations of hMLHl and hMSH2 in China.Methods1. Hereditary predisposition of CRC patients and prevalence of HNPCC in China. Hereditary predisposition of malignant tumors (including CRC) was analyzed in 594 CRC patients who were from Beijing and its neighbor regions. At the same time, documents on HNPCC prevalence in China were chosen, which were publicly published in Chinese or foreign journals. At last, there were accumulatively 4190 patients with CRC together with observed patients by us in all. According to the principle of meta-analysis, we analyzed the prevalence of HNPCC diagnosed by AC I , AC II and JC, respectively.2. Clinical phenotype and screening strategies of HNPCC syndrome in China. Subjects were HNPCC families diagnosed by AC I or AC II. We reviewed the literature on clinical phenotype of HNPCC syndrome in China, and collected AC I or AC II families which were reported. Combined with HNPCC families collected by ourselves, general clinic features were analyzed in 57 AC I or AC II families with complete data, and the spectrum of extracolonic tumors were analyzed in 142 AC I or AC II families with complete data of extracolonic tumors.3. To explore the quantitative method to describing the inner characteristics of HNPCC family. Subjects were 57 AC I or AC II families with complete data. In each HNPCC family, the following parameters were analyzed: median ages diagnosed both the first tumors and the first CRC in patients with HNPCC-associated tumors, proportion of CRC in all first tumors and proportion of CRC patients in all patients with HNPCC-assnciated tumors. At the same time,the common features of HNPCC families were described by these quantitative parameters.4. Establishment and evaluation of quick screening platform of hMLHl and hMSH2 mutation by DHPLC. To amplify the exons of hMLHl and hMSH2, a set of primers were designed by ourselves, which were fit for PCR amplification and DHPLC detection. According to primary structure of each amplicon, we used WAVEMAKER to analyze and choose oven temperatures used by DHPLC in screening mutation of exons and splicing regions. Then in a single blind procedure, special messenger operated the DHPLC and screened hMLHl and hMSH2 mutation in the probands in 26 HNPCC families, who had been detected by DNA sequencing. According to the results of DNA sequencing, sensitivity and specificity of DHPLC were evaluated at last.5. The status of hMLHl and hMSH2 mutation in members of HNPCC families. In 14 HNPCC families with germline hMLHl or hMSH2 mutation, 191 members were detected the status of germline hMLHl and hMSH2mutations by DHPLC, who previously signed the imformed consents. When we screened carriers in each family, the positive control was the proband in this family and the negative control was one of the other probands in the other families who didn't carry the mutation of the same exon of the same gene. In addition, the correlation was analyzed between the status of hMLHl and hMSH2 mutation and the degree of relatives with HNPCC-related tumors.6. Predictive role in cancer risk and genetic features of mismatch repair genes mutation in HNPCC families. 191 members who have a clear status of hMLHl or hMSH2 mutation were followed up in 14 HNPCC families. And coloscopy was finished by special manager in some of them. Based on clinical information of them including 31 putative carriers, cumulative risk of all HNPCC-related tumors, CRC and gastric cancer at one age was calculated using Kaplan-Meier survival analyzed method. Differences of mean cumulative risk were analyzed by Cox's proportional hazard model and log-rank test between hMLHl and hMSH2, man and female.Results1. In 594 patients with CRC, hereditary predisposition of both malignant tumors and CRC were 17.2 % and 5.2%, respectively. Multiple primary cancers and multiple CRCs were 13.1% and 10.1%, respectively. Among them, multiple primary cancers and multiple CRCs without familial history were 9.1 % and 6.7 %, respectively. There was a correlation between familial history of malignant tumors and multiple primary cancers and multiple CRCs (both P< 0.01). Patients with familial history of malignant tumors had more frequently multiple primary cancers and multiple CRCs than patients without history of malignant tumors. The median age diagnosed CRC was 65 years. Younger patients (=50) were 21.4 % in all CRC patients. There was a correlation between younger patients (=50) and familial history of malignant tumors and familial history of CRC (both P< 0.05). The age diagnosed CRC was younger in patients with familial history of malignant tumors and familial history than patients without them. In the population of CRC patients in China, prevalence of HNPCC diagnosed AC I , AC II and JC were 1.24 %, 2.15 % and 2.93 %, respectively.2. HNPCC families fulfilled AC I were mostly Lynch syndrome II. In AC I HNPCC families, the proportion of patients with CRC, extracolonic cancers and multiple primary cancers were 82.4 %, 25.7 % and 19.2 % in all patients with HNPCC-related tumors, respectively. Extracolonic cancers were the only cancers in 18.0% of them. Right colon cancer was 66.8 % in all patients whose first cancer were CRC. Right colon cancer, rectal cancer and multiple CRCs were 66.5 %, 20.5 % and 19.1 % in all patients with CRC, respectively. Normal distribution was found in the distribution of the age diagnosed the first cancer, but not found in the distribution of the age diagnosed the first CRC. When patients diagnosed the first cancer, the mean age and median age were 40.0 years and 45.1 years, respectively. When diagnosed the first CRC, the median age was 43 years. The similar results were drawn in AC II HNPCC families. The onset age of cancer was increasingly younger generation by generation. And specificity of the site of tumors was becoming more obvious when more generations were affected. In China, the spectrum of extracolonic cancers was much wider and included cancers of more sites. Gastric cancer was the most frequent extracolonic cancer, followed by endometrial cancer andhepatocarcinoma.3. FMA-DFMT is the median age at which patients with HNPCC-related tumors were diagnosed the first malignant tumor in each HNPCC families. In these HNPCC families, most of FMA-DFMTs were from 35 years to 50 years, and its median was 44.0 years. FMA-DFCRC is the median age at which patients with HNPCC-related tumors were diagnosed the first CRC in each HNPCC families. The median of FMA-DFCRC was 43.5 years in these HNPCC families. FPCRC-FMT is the proportion of CRC in all the first tumors of all patients with HNPCC-related tumors in each HNPCC family. FMA-DFMT was less than 0.6 in 13.2 % families, and was equal to 1.0 in 56.6 % families. FPP-CRC is the proportion of CRC patients in all patients with HNPCC-related tumors, which also is the parameter described the specificity of onset site of tumor. In addition, there are other parameters which can describe the stability and specificity of clinical phenotype of HNPCC, such as affected generations, the type of family and the number of patients with HNPCC-related tumors and so on.4. A set of primers were designed by ourselves, which can amplify all exons of hMLHl and hMSH2 very well and have a high specificity. A set of oven temperatures was designed by ourselves, which can theoretically detect all mutations in any site of exon regions and splicing regions of hMLHl and hMSH2 very well. When these oven temperatures were used, DHPLC detected all hMLHl and hMSH2 mutations which were previously detected by DNA sequencing. Both false positive and negative results weren't found. The sensitivity and specificity were both 100 % when DHPLC screened hMLHl and hMSH2mutations. But there was significant differences between oven temperatures of exon region and splicing region in the amplicon 12A of hMLHl and the amplicon 2, 3, 7 and 5 of hMSH2 (2.2°C~8.5°C). If mutations are screened in the exon region and splicing region of these amplicons by DHPLC, some mutation may be missed. Compared to DNA sequencing, DHPLC has many merits such as convenient, quick, high efficient, low laborious, low cost, small artificial error and high sensitivity and specificity when DHPLC screens the mutations of the exon region and splicing region of hMLHl and hMSH2.5. In 191 members of IINPCC families, there were 63 carriers (33.0%). Thedetected rate of carrier was 52.5% (62/118). The distribution of carriers was not well-proportioned in members with different ages in HNPCC families. Some patients entailed hMLHl or hMSH2 mutation on less offspring. In some patients and families, there were less or not any offspring with hMLHl or hMSH2 now than before. There was a correlation between the carriers of mutation and the degree of relatives with HNPCC-related tumor. Carriers were almost the members with first-degree relatives with HNPCC-related tumor (P<0.00l). There were few carriers in the members with second-degree relatives with HNPCC-related tumor and not carriers in the members with third or more than degree relatives with HNPCC-related tumor. In the members with third or more than degree relatives with HNPCC-related tumors, detected rate of carriers was correlated with the age of members. The number of patients and siblings were indirectly parameters reflecting the stability of clinical phenotype of HNPCC syndrome.6. Family 4 had a germline hMSH2 splicing mutation (exonl4, c.2279-2A> C). All patients were men in this family, and man carrier was not found not to suffer from tumors in his lifetime. So man carriers were inherited in an autosomal dominant manner. But the female carriers were different from man carriers. Patients with HNPCC-related tumors were not found in female carriers. Among 4 female carriers in this family, one hadn't suffered from tumors in her lifetime (died at 81); the others were currently more than 40 years and were all healthy. It showed female carriers were inherited in an autosomal recessive manner. There were man and female patients with HNPCC-related tumors in the other families. Patients were not found in non-carriers. Carriers were inherited in an autosomal dominant manner in these families. Receptive rate of coloscopy was only 37.1 % in 14 HNPCC families, which wasn't associated with the fact whether they carried an hMLHl or hMSH2 mutation. Adenoma was detected in 30.3% of carriers, but wasn't found in non-carriers. The difference between them was statistically significant (P<0.01). The morbidity of HNPCC-related tumors was 63.8 % in carriers who were currently alive and whose ages were more than 20 years. The relative risk of carriers was 317.6 times than non-carriers in developing HNPCC-related lumors. Among members withfirst-degree relatives with HNPCC-related tumors, the relative risk of carriers was 161.6 times than non-carriers in developing HNPCC-related tumors. At age 30, 40, 50 and 60 years in hMLHl or hMSH2 mutation carriers, the mean cumulative risk of developing a cancer at any site were 9.7 %, 38.9 %, 69.5 % and 92.4 %, respectively; the mean cumulative risk of developing CRC were 9.7 %, 36.4 %, 66.7 % and 81.3 %, respectively; the mean cumulative risk of developing 0, 1.4 %, 6.1 % and 29.6 %, respectively. The mean cumulative risks were increasing when carriers were becoming older. The differences of the mean cumulative risks between hMLHl mutation carriers and hMSH2 mutation carriers had all no statistically significance in developing all HNPCC-related tumors, CRC and gastric cancer (P>0.05). But the difference of the mean cumulative risks between man carriers and female carriers had all statistically significance in developing all HNPCC-related tumors and CRC (i><0.01). When man carriers and female carriers were the same age, the man carriers had a higher risk than the female ones. But when they were 60 years, their risk would reach the peak and have the same risk. When the carriers had the same sex, the difference of the mean cumulative risks between hMLHl mutation carriers and hMSH2 mutation carriers had all no statistically significance in developing all HNPCC-related tumors and CRC (P>0.05).Conclusions1. Hereditary predispositions of malignant tumors and CRC were 17.2 % and 5.2 % in CRC patients, respectively. Familial history of malignant tumors, multiple primary cancers and younger patients with CRC (=50) were the clinical markers of hereditary predisposition of malignant tumors including CRC. Multiple primary cancers and younger patients with CRC (=50) were all correlated with familial history of malignant tumors. The prevalence of HNPCC in China was similar to that in Western countries.2. Clinical phenotype of HNPCC syndrome was similar to those in the other countries. But the spectrum of extracolonic cancers in China was wider and involved with more organs than that in Western countries. Gastric cancer wasthe most frequent extracolonic cancer. The onset age of cancer was younger generation by generation.3. The features of HNPCC families could be accurately described by a series of quantitative parameters.4. A quick platform of screening hMLHl and hMSH2 mutation by DHPLC was successfully established and had very higher sensitivity and specificity. Mutations occurred in amplicon 12A of hMLHl and amplicon 2, 3, 5 and 7 of hMSH2 were suggested to be screening by DNA sequencing.5. For members in HNPCC families, whether they carry hMLHl or hMSH2 mutation is correlated with the degree of their relatives with HNPCC-related tumors. Carriers are almost the members with first-degree relatives with HNPCC-related tumors. Genetic test should be including all members with first-degree relatives with HNPCC-related tumors.6. The heritance of germline mismatch repair genes mutation is very complex and heterogeneous. In hMLHl and hMSH2 mutation carriers, the detected rate of adenoma was very high than in non-carriers; the mean cumulative risk developing HNPCC-related tumors in China was similar to that in Western countries. But the onset age of the Chinese was younger that of the Western. Gastric cancer was the most frequent extracolonic cancer in hMLHl or hMSH2 mutation carriers. The onset age of man carriers was younger than that of female carriers. The difference of the mean cumulative risks between hMLHl mutation carriers and hMSH2 mutation carriers had all no statistically significance in developing all HNPCC-related tumors, CRC and gastric cancer.
|
PDF Full Text Request