Clinical Value of PET.CT Based on Big Data in Colorectal and Peritoneal Metastatic Cancer.

This study focuses on the evaluation of the clinical utility of PET-CT imaging in peritoneal metastases and colorectal cancer. One hundred patients with colorectal peritoneal metastases, who underwent whole-body PET-CT imaging from January 2015 to December 2019, were selected as the experimental group, and 20 healthy individuals were selected as the control group. The SUVmax of the two groups of patients was 5.73 ± 3.84 and 2.70 ± 2.32, respectively, and the difference was statistically significant. The SUVmax AUC was 0.720, and the AUC of serum AFP, CEA, CA125, and CA199 were 0.596, 0.677, 0.642, and 0.696, respectively. Conclusion. 100 patients with colorectal and peritoneal metastatic cancer underwent PET/CT examination. The follow-up or other imaging examinations confirmed the diagnosis. Analysis of the ROC curve in this study found that with a peritoneal SUVmax> 3.2 as the diagnostic index for colorectal peritoneal metastatic cancer, the sum of sensitivity and specificity reached the maximum.

1. Introduction

Early diagnosis of colorectal cancer with peritoneal metastasis is difficult. The mechanism of peritoneal metastasis may be that the tumor breaks through the serous layer or the serous layer is opened during surgery. Due to the small volume and low volume density of peritoneal nodules, early diagnosis is very difficult.

Recent data support the use of intraperitoneal chemotherapy for the treatment of celiac disease [1]. In [2], the authors conducted a systematic search of the library's electronic databases using the term colorectal cancer. In [3], the authors treated 3 impressive patients with colorectal cancer who developed a peritoneal recurrence and underwent several surgeries. In [4], the authors developed dual-targeted nanoparticles with good gene transfection efficiency for gene therapy of peritoneal metastases from colorectal cancer. In [5, 6], the authors underwent partial mitral valve resection and direct closure. In [7], the authors used two cell lines and samples and a combined expression analysis of 200 patients with advanced gastric cancer to determine the drivers of peritoneal spread [8]. In [9], the authors analyzed the transfer pathway-specific transcriptome. In [10], the authors described the experience of four patients with peritoneal disseminated hepatocellular carcinoma (HCC) who were controlled and managed by cytoreductive surgery (CRS) and hot intraperitoneal chemotherapy (HIPEC) [11]. In [12, 13], the authors revealed whether collagen triple helix repeats are a predictor of peritoneal and lymph node metastasis in epithelial ovarian cancer. In [14], the authors built a predictive model and surveyed 1720 patients with stage 1–3 colon cancer. A routine laboratory examination of ascites can determine whether the ascites are exudate or leaking. It has only preliminary hints and cannot be clear in nature.

As a consumer, if the patient cannot be treated by surgery because of their physical condition, chemotherapy or targeted therapy can also be used to control the disease. In addition, patients should also pay attention to their own psychological adjustment, try to relax, actively cooperate with doctors for treatment, and enhance nutrition in their diet. Ent of metabolic intratumoral heterogeneity (ITH) by means of texture features (TF) on 18F-FDG PET/CT alone has the prognostic ability for a variety of tumors [15]. FDG PET/CT is a sensitive diagnostic technique that helps diagnose fever of unknown origin (FUO) [16]. In [17, 18], the authors found that PSMA PET/CT can observe residual lesions or recurrence of prostate cancer at lower PSA levels. In [19], the authors explored the value of 68Ga-PSMA PET/CT in assessing docetaxel response in metastatic castration-resistant prostate cancer. In [20], the authors found that in the treatment of pediatric Hodgkin's lymphoma (HL), temporary PET/CT has better specificity, and the use of Deauville criteria can further improve specificity.

PET/CT has better recognition and diagnostic value for distant organ metastasis including colorectal cancer with peritoneal metastasis. The SUVmax of the two groups of patients was 5.73 ± 3.84 and 2.70 ± 2.32. In this study, the SUVmax AUC was 0.720, and the serum AFP, CEA, CA125, and CA199AUC were 0.596, 0.677, 0.642, and 0.696, respectively. The SUVmax AUC is the largest, and it can be considered that among these research indicators, SUVmax has the highest diagnostic accuracy. Analysis of the ROC curve in this study found that with a peritoneal SUVmax> 3.2 as the diagnostic index for colorectal peritoneal metastatic cancer, the sum of sensitivity and specificity reached the maximum.

2. Method

2.1. Source

2.1.1. The Control Group

Twenty healthy people who came to our hospital for PET-CT central medical examination without complaint were selected. Their physical examination, chest radiograph, abdominal ultrasound, and laboratory tests were normal. There were 10 males and 10 females, aged 35 to 69 (50 ± 7) years.

2.1.2. The Case Group

There were 60 males and 40 females, aged 29 to 77 (56 ± 11) years. The statistical objects must meet the following requirements: (1) the primary site of the tumor is located in the rectum or colon, and the relevant information is complete; (2) the primary tumors of colorectal cancer are biopsied or surgically removed, and the pathological diagnosis is complete; (3) determining the site of peritoneal metastasis requires one or more of the following methods: pathological biopsy, enhanced CT, MRI, PET/CT, bone scan, liver, gallbladder, pancreas, and spleen color Doppler ultrasound; and (4). exclude previous or combined primary malignant tumors at other sites.

The statistical contents include the patient's age, gender, whether it has colorectal peritoneal metastasis, whether there are multiple peritoneal metastasis, the primary tumor site, the levels of tumor markers CEA and CA19-9, and the extraperitoneal metastasis site, etc. In addition, we also counted whether the patient had a PET/CT examination. Relevant medical records were obtained by querying pathological results, imaging, and laboratory test reports.

2.2. Instruments and Reagents

2.2.1. Equipment

The equipment used is the following: Germany Siemen's company Biography 16 sensation imager, PET: advance PET scanner, 39-ring detector, and CT: light speed 16-row spiral CT.

RDS-Eclipse ST cyclotron from Siemens, Germany; Explora FDG4 synthesis module from Siemens, Germany; TLC system from Bioscan, USA; HPLC from Grace Alltech, USA; Flow-Count Radio-HPLC detector system from American Bioscan Company; and Capintec Corporation CRC-15R/PET activity meter.

2.2.2. Main Reagents

Acetonitrile, potassium carbonate, K22, and HCl from Aldrich, USA; sterile injection water from Baxter, USA; QMA, C-18, ALU column from waters, USA; and ag11a8 and ag50 w purified resin from Bio-Rad, USA.

2.3. PET.CT Imaging Methods and Processing

The basic principle of PET-CT is to use PET and CT combined imaging, by introducing radionuclides for imaging, and then using CT anatomy for combined diagnosis. The imaging agents mainly introduced in the imaging include metabolites, glucose, amino acids, proteins, and polypeptides and other elements, which belong to the comprehensive molecular imaging technology.

The patient fasted for more than 6 hours, did not do strenuous exercise, and the peripheral blood glucose was controlled at (3.1∼11.1) mmol/L. In a calm state, the imaging agent (3.7∼5.5) MBq/kg was intravenously injected according to the body weight, and the patient was rested in the room for 45 min ∼1 h. After the bladder was emptied, a PET-CT whole-body imaging was performed. All patients received an oral mass fraction of 2% diatrizoate 600 ml 40 minutes before the examination, and 2% diatrizoate 600 ml orally 5 minutes before the examination. CT scanning conditions included the following:voltage, 120 kV; current, 100 mA; pitch, 1; the pitch of single turn of the tube, 0.3 s; and layer thickness, 5 mm, and PET scanning conditions were as follows:PET images were collected in three dimensions in the same range. Generally, 7 to 8 beds were used. PET scans were performed on the body at 1.8 min/bed, and 4.5 min/bed in the brain. During the whole-body imaging, the patient breathed calmly and evenly so that the CT and PET images matched.

2.4. Evaluation Criteria

It refers to the radioactivity of the imaging agent absorbed by the local tissue and the average injection activity of the whole body, the evaluation content of the PET-CT system evaluation standard, and the evaluation content of the index application materials. The evaluation index function is located in the radiotherapy qualification and practice license approved by the health administrative department and the tumor-related department.

PET-CT diagnostic criteria for peritoneal metastasis include the following:

Limited peritoneal radiation uptake and CT on the corresponding site showed nodular thickening of the peritoneum;

Localized or diffuse radioactive uptake of the peritoneum was increased, and no obvious abnormalities were found on the CT at the corresponding site when combined with other examination data of the patient for diagnosis;

No radioactive uptake was found in the peritoneum. CT showed peritoneal lesions and met the CT diagnostic criteria.

The final clinical diagnosis was compared with the PET-CT scan data.

True positive (TP): PET-CT is diagnosed as malignant, and it is considered as true positive if it is consistent with the clinical diagnosis; true positive rate: sensitivity and the ratio of true positive cases of PET-CT in the group of cases diagnosed as malignant by the gold standard (%);

False positive (FP): PET-CT is diagnosed as malignant and is clinically diagnosed as benign, then, it is false positive; false-positive rate: the rate of misdiagnosis and the ratio of the number of PET-CT false-positive cases in the case group diagnosed as benign by the gold standard (%);

True negative (TN): PET-CT diagnosis is benign, and it is considered as true negative if it is consistent with clinical diagnosis; true negative rate: the specificity and the ratio of the number of true negative cases of PET-CT in the case group diagnosed as benign by the gold standard (%);

False negative (FN): PET-CT is diagnosed as benign and clinically diagnosed as malignant, then it is considered to be false negative; false negative rate: the rate of missed diagnosis and the ratio of the number of PET-CT false negative cases in the group of cases diagnosed as malignant by gold (%).

Accuracy rate: the sum of the true negative diagnoses of PET-CT and the proportion (%) of the total number of cases.

3. Results

3.1. Analysis of Tumor Metastasis Patterns in Colorectal and Peritoneal Metastatic Cancer

The age ranged from 29 to 77 (56 ± 11) years. There were 37 patients with peritoneal metastasis of simultaneous colon cancer, accounting for 82.2%, and 33 cases of peritoneal metastasis of simultaneous rectal cancer, accounting for 60%. The proportion is higher, that is, 83.6% and 84.4%, respectively. Patients with rectal cancer and colon cancer are more likely to have peritoneal metastasis, and there is no statistical difference between them (p > 0.05, Table 1). The 21 patients had peritoneal and extraperitoneal metastases; colon cancer patients with peritoneal and extraperitoneal metastases were 28 cases, accounting for 62.2%. Compared with rectal cancer patients, colon cancer patients were more likely to have extraperitoneal metastases. Analysis of tumor metastasis patterns will not only have an impact on the understanding of tumor metastasis evolution but will also have important clinical implications. Tumor cells are dormant until they acquire the full metastatic capacity to grow unrestricted, and metastases can be established when the primary is small or undetectable (i.e., metastases of unknown primary). Possible reasons may be that it is the early disseminated tumor cells that establish a favorable microenvironment for growth in distant sites, and the growth rate exceeds that of the primary tumor.

Table 1

Tumor metastasis patterns of colorectal peritoneal metastatic cancer.

	Number of cases (n)	Simultaneous peritoneal metastasis (n/%)	Multiple peritoneal metastases (n/%)	Combined peritoneal metastases (n/%)
Rectal cancer	55	33 (60.0%)	46 (83.6%)	21 (38.2%)
Colon cancer	45	37 (82.2%)	38 (84.4%)	28 (62.2%)
Statistical values	—	4.291	3.067	4.892
P	—	0.164	0.283	0.061

The ways of peritoneal metastasis of colorectal cancer mainly include the following two aspects: (1) tumor cells break through the serosa and then fall off into the abdominal cavity, and then grow further in the peritoneum; and (2) Iatrogenic factors, severed blood vessels, and lymphatic tumor emboli follow the blood flow. The inflow of lymph and lymph into the abdominal cavity, the pulling and squeezing of the tumor tissue during the operation, and the inflow of tumor cells into the abdominal cavity through the intestinal stump with the intestinal fluid can all lead to the implantation of tumor cells in the abdominal cavity during the operation. A total of 49 patients in the two groups were combined with extraperitoneal metastases. The average age of those with peritoneal and extraperitoneal metastases was (48.12 ± 6.33) years. The most common site of extraperitoneal tumor metastasis in 49 patients was lung metastasis, and colorectal and rectal cancer with peritoneal and lung metastases were all above 65%, although a larger proportion of patients in the colon cancer group showed lung metastases (71.4%). However, there was no statistical difference in patients with colon and rectal cancer. The main surgical method for peritoneal metastases is cytoreductive surgery. The purpose of surgery is to remove all visible tumor nodules located in the parietal and visceral peritoneum as much as possible. There were 32 patients with distant metastasis in two or more extraperitoneal sites, including 13 in the rectal cancer group (61.9%) and 19 in the colorectal cancer group (67.9%), as shown in Table 2.

Table 2

Analysis of metastatic sites of colorectal peritoneal metastasis.

Extraperitoneal metastases	Rectal cancer (n/%)	Colon cancer (n/%)	Statistical values	P value
Lung	14 (66.7%)	20 (71.4%)	1.602	0.206
Liver	3 (14.3%)	3 (10.7%)	9.604	0.002
Pelvic cavity	1 (4.8%)	2 (7.1%)	0.088	0.767
Bone	2 (9.5%)	0 (0%)	0.242	0.623
Other	1 (4.8%)	3 (10.7%)	3.91	0.048
2 or more places	13 (61.9%)	19 (67.9%)	1.643	0.2
Total patients (n)	21	28	—	—

3.2. Analysis of PET-CT in the Diagnosis of Colorectal and Peritoneal Metastatic Cancer

In all patients with ascites, CT diagnosed 50 cases of malignant ascites by peritoneal morphological changes, 6 of which were false positive, the cause was 4 cases of tuberculous peritonitis, 1 case of liver cirrhosis, and 1 case of benign effusion of unknown cause. Of the 61 patients diagnosed with benign effusion, 35 were false negative. PET-CT can be used to detect occult metastases and lesions that cannot be detected by conventional imaging methods. Several small, uncontrolled trials have previously supported the use of PET-CT in preoperative staging of colorectal liver metastases to avoid unnecessary surgery. However, these pieces of evidence are not sufficiently convincing. The etiology was 14 cases of liver cancer, 5 cases of stomach cancer, 4 cases of colon cancer, and 2 cases of lung and pancreal cancer, as shown in Table 3.

Table 3

Results of PET-CT in the diagnosis of colorectal peritoneal metastasis.

Modality	TP	FN	TN	FP	Sensitivity	Specificity	PPV	NPV	Accuracy
PET/CT	73	3	22	2	85.8%	82.5%	90.33%	80.29%	95.8%
CT	40	19	32	9	70.8%	61.2%	89.72%	52.37%	76.7%
The McNemar test exact Sig. (2-sided)					0.000∗	0.622	0.000∗	0.000∗	0.000∗

Areas under the curve (AUC) of SUVmax, serum AFP, CEA, CAl25, and CAl99 were 0.783, 0.596, 0.671, 0.648, and 0.688, and P values were 0.001, 0.168, 0.009, 0.079, and 0.004 (Figures 1 and 2). It is found in the analysis of the ROC curve that when the SUVmax is 3.2, the sum of the sensitivity and specificity reaches a maximum value of 1.623.

Figure 1

ROC curve of SUVmax, serum AFP, CEA, CA125, and CA199 to identify colorectal peritoneal metastatic cancer.

Figure 2

ROC curve analysis of SUVmax, serum AFP, CEA, CA125, and CA199 to identify colorectal peritoneal metastatic cancer.

3.3. Multivariate and Univariate Analysis of Colorectal and Peritoneal Metastatic Cancer

A total of 49 patients underwent PET/CT examination, of which 22 cases had extraperitoneal metastasis, accounting for 44.9% of the total number of colorectal cancer with peritoneal and extraperitoneal metastasis, and 32 cases had peritoneal metastasis alone. PET/CT in 32 patients with extraperitoneal metastasis confirmed follow-up and other imaging examinations to confirm the diagnosis, and 17 patients with peritoneal metastases alone were misdiagnosed. Follow-up confirmed pulmonary inflammation and was relieved by anti-infective drugs. The specificity and sensitivity of PET/CT in the diagnosis of colorectal peritoneal metastatic cancer were 98% and 96.23%, respectively. The univariate analysis results showed that gender, CEA, CA19-9, and multiple peritoneal metastases were statistically significant with colorectal cancer combined with peritoneal and extraperitoneal metastases. The results of multivariate regression analysis showed CEA, CA19-9, and multiple peritoneal metastases, as shown in Figure 3 and Table 4.

Figure 3

Multivariate and univariate analysis of colorectal and peritoneal metastatic cancer: (a) statistical value and (b) P value.

Table 4

Application of 18F-FDG PET/CT in the diagnosis of colorectal and peritoneal metastatic cancer and extraperitoneal metastasis.

	PET/CT extraperitoneal metastasis (n)	PET/CT peritoneal metastasis (n)	Total (n)
Extraperitoneal metastasis (n)	22	10	32
Peritoneal metastasis (n)	4	13	17
Total (n)	26	23	49

4. Discussion

In the past, most physicians considered PC to be a widespread metastasis of cancer and an end-stage or advanced manifestation of the tumor, and its treatment was often conservative, including systemic chemotherapy, supportive therapy and/or palliative surgery, but the treatment effect was not satisfactory, and the median survival was generally not higher than 6 months. Although the continuous improvement of surgical methods, the constant updating of radiotherapy and chemotherapy drugs, and the emergence of new therapeutic advances such as bioimmunotherapy and targeted therapy have given hope to patients and physicians, the survival rate and remaining survival time of CRC patients and PC patients were still not improved.

Since the 1980s, the international oncology community has considered that focal peritoneal metastasis without distant metastasis is no longer exactly a manifestation of widespread cancer metastasis, but a regional metastasis and active regional treatment measures still have great clinical value, so a set of MTD treatment models was suggested including surgery, bringing obvious survival benefits to patients. This has led to the creation of a set of MTD treatment modalities including surgical, intraoperative, and early postoperative intraperitoneal chemotherapy, which have brought significant survival benefits to patients and are gaining attention.

This ensures that the technology meets the requirements for precision and standardization. HIPEC can achieve a temperature measurement accuracy below 0.1 °C, its temperature accuracy control can be maintained within 0.5 °C, and the peritoneal fluid perfusion rate can be accurate to within 5%. HIPEC can remove ultrafine filtration accuracy below 15 um, and remove free cancer cells and tiny cancer nodules below 3 mm through the scouring force generated by the flow rate of the perfusate.

In this study, a total of 49 patients with colorectal and peritoneal metastases underwent PET/CT examinations, including 22 patients with combined peritoneal and extraperitoneal metastases, and 10 patients with peritoneal metastases alone. Examination of patients with extraperitoneal metastases was confirmed by follow-up and other imaging examinations, and 17 patients with peritoneal metastases alone were misdiagnosed.

The SUVmax of the two groups of patients was 5.73 ± 3.84 and 2.70 ± 2.32, respectively, and the difference was statistically significant. This indicates that most of the peritoneal metastases have increased 18F-FDG uptake on PET-CT imaging, and thus benign and malignant lesions can be judged by metabolism. Due to the large standard deviation, there is still a crossover area between benign and malignant peritoneal lesions SUVmax, which is also the main reason for false negatives and false positives. PET-CT diagnosed 77 cases of malignant ascites, of which 4 cases were false positive, 2 cases were tuberculous peritonitis, 1 case was sclerosing mesenteritis and 1 case was cirrhosis. Malignant peritoneal lesions were diagnosed because of peritoneal miliary and nodular thickening with increased metabolism. Of the 34 patients diagnosed as benign, 6 were false negative. PET-CT also missed the diagnosis of colorectal peritoneal metastatic cancer. The reasons may be as follows: ① well-differentiated tumors have low or no uptake of 18F-FDG, ② tumor cells are in a dormant phase, and ③ it is related to the type of tumor histopathology.

In this article, the SUVmax of PET-CT peritoneal lesions and serum AFP, CEA, CA125, and CA199 were analyzed by the ROC curve, and the diagnostic value of each index was compared by the area under the curve (AUC). The ROC AUC is generally considered to be meaningful between 1.0 and 0.5. When the AUC is equal to 0.5, it indicates that the method is not diagnostic. In this study, the SUVmax AUC was 0.720, and the serum AFP, CEA, CA125, and CA199AUC were 0.596, 0.677, 0.642, and 0.696, respectively. The SUVmax AUC is the largest, and it can be considered that among these research indicators, SUVmax has the highest diagnostic accuracy. Generally, it is believed that SUVmax> 2.5 can diagnose malignant lesions.

5. Conclusions

The specificity and sensitivity of PET/CT for diagnosis of colorectal cancer and peritoneal metastasis combined with extraperitoneal metastasis were 98% and 96.23%, respectively. PET/CT has better recognition and diagnostic value for distant organ metastasis including colorectal cancer with peritoneal metastasis.

PET-CT not only can accurately locate but also has high qualitative value, which helps to identify the nature of ascites. In this study, PET-CT images of 100 patients with colorectal and peritoneal metastatic cancer were analyzed. The SUVmax of the two groups of patients was 5.73 ± 3.84 and 2.70 ± 2.32, respectively. Due to the large standard deviation, there is still a crossover area between benign and malignant peritoneal lesions SUVmax, which is also the main reason for false negatives and false positives.

In this study, the SUVmax AUC was 0.720, and the serum AFP, CEA, CA125, and CA199AUC were 0.596, 0.677, 0.642, and 0.696, respectively. The SUVmax AUC is the largest, and it can be considered that among these research indicators, SUVmax has the highest diagnostic accuracy. Generally, it is believed that SUVmax> 2.5 can diagnose malignant lesions in this study.