Liu XH, Chen G, Cao DD, Liu H, Ke XK, Hu YG, Tan W, Ke D, Xu XM. Comparison of clinical features of patients with or without severe gastrointestinal complications in aggressive gastrointestinal lymphomas. World J Gastrointest Oncol 2024; 16(11): 4409-4423 [PMID: 39554739 DOI: 10.4251/wjgo.v16.i11.4409]
Corresponding Author of This Article
Xi-Ming Xu, Doctor, Chief Physician, Professor, Department of Cancer Center, Renmin Hospital of Wuhan University, No. 238 Jiefang Road, Wuhan 430000, Hubei Province, China. doctorxu120@aliyun.com
Research Domain of This Article
Oncology
Article-Type of This Article
Retrospective Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Xiao-Hong Liu, Gong Chen, De-Dong Cao, Xi-Ming Xu, Department of Cancer Center, Renmin Hospital of Wuhan University, Wuhan 430000, Hubei Province, China
Hui Liu, Department of Hematology, Renmin Hospital of Wuhan University, Wuhan 430000, Hubei Province, China
Xiao-Kang Ke, Department of Pathology, Renmin Hospital of Wuhan University, Wuhan 430000, Hubei Province, China
Yu-Gang Hu, Department of Ultrasound Imaging, Renmin Hospital of Wuhan University, Wuhan 430000, Hubei Province, China
Wei Tan, Department of Gastroenterology, Renmin Hospital of Wuhan University, Wuhan 430000, Hubei Province, China
Dong Ke, Department of Gastrointestinal Surgery, Renmin Hospital of Wuhan University, Wuhan 430000, Hubei Province, China
Author contributions: Liu XH and Liu H designed the study and interpreted the data; Chen G, Ke XK, Ke D, Tan W and Cao DD contributed to the acquisition of the clinical samples and data; Hu YG, Cao DD, Liu XH and Xu XM performed the analytic calculations and wrote the manuscript; All authors have read and approved the final manuscript.
Institutional review board statement: The study was approved by the Medical Ethics Committee of Renmin Hospital of Wuhan University, No. WDRY2021-KS024.
Informed consent statement: Patient consent was waived by the IRB because all patient data was de-identified.
Conflict-of-interest statement: The authors declare that they have no conflict of interest.
Data sharing statement: Pathology data and the statistical analyses for the current study are available from the corresponding author upon reasonable request.
Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Xi-Ming Xu, Doctor, Chief Physician, Professor, Department of Cancer Center, Renmin Hospital of Wuhan University, No. 238 Jiefang Road, Wuhan 430000, Hubei Province, China. doctorxu120@aliyun.com
Received: July 17, 2024 Revised: August 25, 2024 Accepted: September 24, 2024 Published online: November 15, 2024 Processing time: 99 Days and 20.1 Hours
Abstract
BACKGROUND
Aggressive primary gastrointestinal non-Hodgkin lymphoma (PGINHL) is an uncommon and heterogeneous group of lymphoid malignancies, that differs from indolent lymphoma and has a high incidence of severe gastrointestinal complications (GICs).
AIM
To investigate and compare the clinicopathological characteristics, treatments and outcomes in the GICs and No-GICs group with aggressive PGINHL.
METHODS
This retrospective analysis was performed on aggressive PGINHL patients between January 2013 and December 2021 at our hospital. The independent influence factors of GICs were obtained by univariate and multivariate Logistic regression analysis, the selected variables significantly related to GICs were selected as the final predictors to construct nomogram. Kaplan-Meier curves further analyzed the survival of patients in GICs and No-GICs groups. Survival analysis of GICs group was performed using Cox regression.
RESULTS
We focused on 124 aggressive PGINHL cases, which had a relatively high incidence 48.4% (60/124 cases) of GICs, the most common histological type in GICs group was diffuse large B-cell lymphoma (DLBCL) (n = 49, 81.7%). In the GICs group, small intestine was the most common anatomic site of lesion (43.3%), followed by large intestine (31.7%), and then stomach and esophagus (25.0%). Multivariate Logistic regression analysis showed that the independent risk factors for GICs were the small intestine [odd ratio (OR) = 3.33; 95% confidence interval (CI): 1.47-9.41; P = 0.009), aggressive B-cell (OR = 0.09; 95%CI: 0.01-0.83; P = 0.033), maximum tumor diameter (OR = 1.25; 95%CI: 1.07-1.47; P = 0.005), invaded deep serous layer (OR = 3.38; 95%CI: 1.24-9.19; P = 0.017). We developed a nomogram to predict risk of GICs in aggressive PGINHL patients based on independent risk factors. The value of area under curve calculated by receiver operating characteristic curve was 0.815, and calibration curve and decision curve analysis further indicated that the prediction effect was superior. The majority of patients with GICs were given combination therapy (chemotherapy combined with surgery or radiation). Event-free survival and overall survival in GICs group were no worse than those in the No-GICs group.
CONCLUSION
The complication rate of GICs in patients with aggressive PGINHL was relatively high, particularly in PGI-DLBCL. The independent risk factors for GICs were the small intestine, PGI-TNKL, bulky tumor, and depth of invasion. A combination treatment, involving surgery, improved survival in the GICs group.
Core Tip: The characteristics of aggressive primary gastrointestinal non-Hodgkin lymphoma (PGINHL) with or without severe gastrointestinal complications (GICs) were explored in this study. The relationship between GICs and clinicopathological features of aggressive PGINHL patients, such as primary site, histological type, tumor size, depth of gastrointestinal invasion and other factors, was investigated by statistical analysis. The aim of this study was to summarize the risk factors and treatment strategies for aggressive PGINHL patients with or without GICs and to provide evidence-based evidence for clinical decision-making and individualized treatment strategies.
Citation: Liu XH, Chen G, Cao DD, Liu H, Ke XK, Hu YG, Tan W, Ke D, Xu XM. Comparison of clinical features of patients with or without severe gastrointestinal complications in aggressive gastrointestinal lymphomas. World J Gastrointest Oncol 2024; 16(11): 4409-4423
Primary extra-nodal lymphoma was defined as a lymphoma whose primary site was not the lymph node but originated in tissues and organs other than the lymph node[1]. The digestive system organs had relatively rich lymphatic tissue, especially the gastrointestinal tract lymphatic drainage area, so the incidence of lymphoma invading the gastrointestinal tract was relatively high. Primary gastrointestinal lymphomas were rare compared to epithelial cancers, accounting for only 1% of gastrointestinal tumors. The most commonly involved organ was the stomach, followed by the small and large intestine, while the esophagus, appendix, and rectum were rare[2]. Primary gastrointestinal lymphoma was classified into B cell and nature killer (NK)/T cell lymphoma. It also was divided into aggressive and indolent lymphoma according to the course of the disease. Indolent B-cell lymphoma such as mucosal associated lymphoid tissue (MALT) and follicular cell lymphoma (FL) had little invasive damage to the gastrointestinal tract[1,2]. The indolent T-cell and NK-cell lymphomas usually had non-destructive infiltration. Among aggressive primary gastrointestinal non-Hodgkin lymphoma (PGINHL), diffuse large B-cell lymphoma (DLBCL) was the most common pathological subtype (47.4%)[3]. Primary gastrointestinal T/NK cell lymphomas (PGI-TNKL) were also aggressive entities consisting of various subtypes with distinct clinicopathological features and prognoses[4-6]. Aggressive gastrointestinal lymphoma had a more invasive growth and a worse prognosis than indolent lymphoma. Due to aggressive lymphoma occurred in the gastrointestinal tract, aggressive behaviors such as severe gastrointestinal complications (GICs)-gastrointestinal bleeding (GIB), gastrointestinal obstruction (GIO), and gastrointestinal perforation (GIP) were also common at the first visit, especially after chemotherapy, GIP can lead to a very high mortality rate[7-9]. Special attention should be paid to the high risk of GICs in PGINHL patients, necessitating a multidisciplinary discussion and even emergency surgical intervention.
Early identification of GICs risk factors may be beneficial to early treatment and improve the prognosis of aggressive PGINHL patients[1]. A retrospective study suggested that GICs occurred in 56.8% of 148 PGI-DLBCL patients[7]. DLBCL was the most common type of aggressive PGINHL. T-cell lymphomas were uncommon, estimated to account for 4% to 6% of GI tract lymphomas[10]. A relatively large cohort study of PGI-TNKL was performed in 38 patients from South Korea[11]. Aggressive PGI-TNKL mainly referred to as intestinal T-cell lymphoma (ITCL)[1,10]. ITCLs comprised two main entities: Enteropathy-associated T-cell lymphoma (EATL) and monomorphic epitheliotropic intestinal T-cell lymphoma (MEITL), plus ITCL, not specifically (NOS)[5], they were both aggressive in behavior but differ in their clinicpathological features[11]. Aggressive extranodal natural killer/T-cell lymphoma (NKTCL) was more common in Asia, gastrointestinal tract also was involved[12]. However, no studies had systematically summarized the occurrence of GICs in aggressive gastrointestinal lymphomas.
At present, there were few prospective studies on treatments. How to identify high-risk GICs in aggressive PGINHL patients, improving the therapeutic effect and overall survival time was still under discussion. In this study, we analyzed aggressive PGINHL patients in our cancer center in the past 10 years. We were the first to analyze the GICs group vs the No-GICs group in aggressive PGINHL, to find risk factors that predicted GICs, and establish a prediction nomogram, which was conducive to the early identification of clinicians. At the same time, we analyzed the treatment characteristics and prognosis of aggressive PGINHL with GICs, and attempted to provide evidence-based basis for treatment.
MATERIALS AND METHODS
Patient Selection and evaluation criterion
The patients included inpatient and outpatient lymphoma patients at our hospital from January 2013 to December 2021, and followed up until December 2022. Follow-up methods included in-patient review, outpatient review, and telephone follow-up. Primary gastrointestinal lymphoma was defined as predominant lesions in the alimentary tract with or without regional lymph node involvement[13,14]. All cases were diagnosed according to the World Health Organization diagnostic criteria[1]. In our cancer center, the flow chart of the selected aggressive gastrointestinal lymphomas was shown in Figure 1. The aggressive PGI-B-cell lymphomas were DLBCL and Burkitt’s lymphoma (BL). ITCL patients accounted for the majority including aggressive PGI-TNKL, specifically EATL, MEITL, ITCL-NOS, and NKTCL. The exclusion criteria were defined as follows: Indolent primary PGINHL, pathological diagnosis inconsistent, infected with human immunodeficiency virus, younger than 18 years.
The clinical data of all patients at the time of initial diagnosis were collected in our cancer center, including age, gender, disease stage, eastern cooperative oncology group (ECOG) score, blood lactate dehydrogenase (LDH), hepatitis virus, Epstein-Barr virus (EBV) DNA, primary site, immunohistochemical markers, tumor size, endoscopy or endoscopic ultrasonography, imaging, bone marrow examinations, treatment methods, and outcomes. Ann Arbor staging was still the main clinical staging, referring to the modified Lugano staging[15]. This retrospective research was approved by the Medical Ethics Committee of Renmin Hospital of Wuhan University (No. WDRY2021-KS024) and conducted in accordance with the declaration of Helsinki.
Pathological immunophenotypic markers
Diagnosis confirmation depended on the pathology of endoscopic biopsy or surgical specimens. The subtype of each lesion was determined using immunohistochemical analysis. According to the expression of cluster of differentiation (CD) 10, multiple myeloma oncogene 1, and human B-cell lymphoma 6 in Han’s classification, patients were classified as germinal center B cells (GCB) and non-GCB. Double expression was defined as MYC expression ≥ 40% and Bcl-2 expression ≥ 50%[16]. MYC gene rearrangement and lesion positivity for EBV-encoded small RNA were assayed by in situ hybridization.
Macroscopy/endoscopy and imaging findings
Macroscopic findings regarding morphology and size were reviewed by expert endoscopists and pathologists based on the endoscopic findings and/or surgically resected specimens. Lesion morphology was classified as polypoid, ulcerative, diffuse nodular, or diffuse-infiltrating type in some literatures[14,17-20]. In our study, we divided these types into neoplasm, ulceration and other types. The neoplasm consisted of polypoid or mass without surface ulceration. Ulceration included diffuse ulceration or focal/elevated ulceration. Slightly thickened or inconspicuous luminal walls were classified as others. For later statistical analyses, the depth of tumor invasion was classified as inside or outside the serosal layer based on endoscopic ultrasonography or postoperative pathological findings in our cases.
The radiologist’s first impression of imaging was classified as normal, non-specific inflammation, cancer, or lymphomain some literature[14]. In our study, imaging findings were divided into two categories: Neoplastic lesions or others (normal, non-specific inflammation). 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) and maximum standardized uptake value could not be obtained before diagnosis because most patients in the GICs group were diagnosed after acute abdomen surgery.
Evaluation outcome criteria
GICs were confirmed by clinical symptoms, laboratory tests, imaging, and /or surgery. GIO, including partial and complete obstruction, was considered to be the absence of passage of flatus or feces[21]. GIP was defined as the presence of free air under the diaphragm on abdominal imaging, or intestinal perforation during laparotomy[22]. GIB was defined as overt bleeding (hematemesis, bloody stools) and drop in hemoglobin of ≥ 2 g/dL[23]. All the above GICs were confirmed using imaging or surgery. During the course of the disease, it was possible to develop two or more complications. Patients with GICs required surgery, we also included those who suffered from GICs but could not undergo surgery due to their poor general condition.
Revised response criteria, known as the 2014 Lugano criteria, was divided into imaging remission and metabolic remission using FDG-PET scans[15]. The response was categorized as complete remission, partial remission, stable disease and relapsed disease or progressive disease. Event-free survival (EFS) was defined as the time from the diagnosis to disease progression, recurrence, death or last follow-up. The overall survival (OS) duration was measured from the time of diagnosis to the time of death or last follow-up.
Statistical analysis
After data collection, frequency (%) was used as a descriptive statistic. Quantitative variables were expressed as mean ± SD or median (quartile range). The differences in clinical parameters were compared using χ2 test or Fisher’s exact for categorical variables, or one-way analysis of variance for continuous variables. P values < 0.05 at both sides were considered statistically significant. The GICs risk nomogram was constructed with the most significant factors associated with GICs using Logistic regression analysis. The receiver operating characteristic (ROC) curve was used to evaluate the accuracy of nomogram in predicting the risk of GICs. Then, calibration curves were generated for the comparison between the actual outcomes and nomogram-predicted outcomes. Finally, decision curve analysis (DCA) was conducted by measuring the net benefits for a group of threshold probabilities to measure clinical utility[24]. EFS and OS were evaluated using Kaplan Meier analysis with a log rank comparison. Cox regression univariate and multivariate analyses were used to assess predictors of survival. Statistical analyses were performed using statistical product and service solutions (SPSS) software (version 25.0) for Windows (SPSS Inc., Chicago, IL, United States), R software (version 3.3.0) (http://www.r-project.org/).
RESULTS
Clinicopathologic characteristics of GICs group with aggressive PGINHL
Among approximately 224 PGINHL cases diagnosed from 2013 to 2021, more than 60 indolent gastrointestinal lymphomas (MALT, FL) were excluded, ten cases of mantle cell lymphoma (MCL) with small B-cell lymphoma were also excluded, leaving 124 patients with aggressive PGINHL. Among them, aggressive PGI-B cell lymphomas accounted for 90.3% (110 cases of DLBCL and 2 cases of BL), aggressive PGI-TNKLs only accounted for 9.7% (11 cases of ITCL and 1 case of NKTCL). The final follow-up period was December 2022 (Figure 1).
In our study of 124 cases of aggressive PGINHL, GICs occurred in 48.4% (60/124), including 36 cases of GIO, 9 cases of GIP, 12 cases of GIB, and 3 cases of GIO and GIB. In the subgroup that combined GICs, the median age was 61 years (range: 28 to 89 years), and the male to female ratio was 1.73:1. The most common histological type in the GICs group was DLBCL (n = 49, 81.7%), followed by ITCL (n = 10, 16.7%), NKTCL (n = 1, 1.6%). The most frequently affected anatomic location in GICs group was the small intestine (43.3%), followed by large intestine (31.7%), stomach and esophagus (25.0%), and tumor maximum diameter was 7.03 cm ± 3.7 cm. Compared with the No-GICs group (8/37 cases), the GIC group had a significantly higher proportion of serosal layer and outside (29/37 cases). No matter in No-GICs group or GICs group, in terms of the macroscopic appearance of gastrointestinal tract growth (through endoscopy or surgery), lymphoma was mainly neoplasm, followed by an ulcerative type. In terms of imaging findings, radiologists often considered neoplastic lesions, but they could not further indicate cancer or lymphoma (Table 1). Supplementary Figure 1 showed an endoscopic picture of inflammatory changes in a patient with primary small intestinal TNKL, and another endoscopic picture of a large ulcer in a patient with primary ileocecal junction DLBCL was shown in Supplementary Figure 2.
Table 1 Clinicopathologic characteristics of 124 cases of aggressive primary gastrointestinal non-Hodgkin lymphoma, n (%).
Characteristics
No-GICs (n = 64)
GICs (n = 60)
P value
Age, mean ± SD
59.8 ± 12.5
60.0 ± 14.2
0.932
≥ 60
32 (50.0)
34 (56.7)
0.573
< 60
32 (50.0)
26 (43.3)
Gender
0.427
Male
35 (54.7)
38 (63.3)
Female
29 (45.3)
22 (36.7)
Stage
0.875
I/II
29 (45.3)
29 (48.3)
III/IV
35 (54.7)
31 (51.7)
LDH
0.406
Elevated
27 (42.2)
20 (33.3)
Normal
37 (57.8)
40 (66.7)
Virus
0.876
Yes
8 (12.5)
6 (10.0)
No
56 (87.5)
54 (90.0)
ECOG
0.139
0-1
59 (92.2)
49 (41.7)
≥ 2
5 (7.8)
11 (18.3)
Depth
< 0.001
Inside serous membrane
54 (87.5)
31 (51.7)
Serosal layer and outside
8 (12.5)
29 (48.3)
Diameter, mean ± SD
4.65 ± 2.4
7.03 ± 3.7
< 0.001
Primary site
< 0.001
Esophagus or stomach
40 (62.5)
15 (25.0)
Small intestine
12 (18.8)
26 (43.3)
Colon
12 (18.8)
19 (31.7)
Histopathology
< 0.001
DLBCL + BL
63 (98.4)
49 (81.7)
ITCL + NKTCL
1 (1.6)
11 (18.3)
N-GCB
39 (60.9)
43 (71.7)
0.284
CD5 +
13 (20.3)
7 (11.7)
0.287
CD30 +
5 (7.8)
7 (11.7)
0.673
Double expression
18 (28.1)
9 (15.0)
0.121
Ki 67 ≥ 70%
39 (60.9)
40 (66.7)
0.634
Macroscopic findings
0.858
Neoplasm
29 (45.3)
30 (50.0)
Ulceration
31 (48.4)
27 (45.0)
Others
4 (6.2)
3 (5.0)
Imaging findings
0.205
Neoplasticlesions
53 (82.8)
43 (71.7)
Others
11 (17.2)
17 (28.3)
Risk factors associated with GICs in aggressive PGINHL patients
Results of univariate analysis [odds ratio (OR); 95% confidence interval (CI); P value] were as follows: Primary site (small intestine OR = 6.55; 95%CI: 2.61-16.40; P < 0.001; colon OR = 5.60; 95%CI: 2.15-14.61; P < 0.001), aggressive B cell lymphomas (OR = 0.09; 95%CI: 0.01-0.83; P = 0.033), max tumor diameter (OR = 1.30; 95%CI: 1.13-1.48; P < 0.001), invasive depth (serosal layer and outside OR = 6.55; 95%CI: 2.67-16.06; P < 0.001), ECOG ≥ 2 (OR = 3.35; 95%CI: 1.02-11.05; P = 0.047), and double expression (OR = 0.40; 95%CI: 0.16-0.97; P = 0.044) were risk factors for GICs.
Multivariate analysis revealed that primary site (small intestine OR = 3.33; 95%CI: 1.47-9.41; P = 0.009; colon OR = 2.50; 95%CI: 1.17-7.15; P = 0.023), aggressive B cell lymphomas (OR = 0.09; 95%CI: 0.01-0.83; P = 0.033), max tumor diameter (OR = 1.25; 95%CI: 1.07-1.47; P = 0.005), invasive depth (serosal layer and outside OR = 3.38; 95%CI: 1.24-9.19; P = 0.017) were independent indicators for GICs (Table 2).
Table 2 Univariate and multivariate analyses of factors associated with gastrointestinal complications in aggressive Primary gastrointestinal non-Hodgkin lymphomas patients.
Characteristics
Univariate analysis
Multivariate analysis
OR (95%CI)
P value
OR (95%CI)
P value
Age
≥ 60
1.21 (0.60-2.45)
0.597
< 60
Ref.
-
Gender
Male
1.48 (0.72-3.03)
0.289
Female
Ref.
-
Primary site
Esophagus or stomach
Ref.
-
Ref.
Small intestine
6.55 (2.61-16.40)
< 0.001
3.33 (1.47-9.41)
0.009
Colon
5.60 (2.15-14.61)
< 0.001
2.50 (1.17-7.15)
0.023
Histopathology
DLBCL + BL
0.08 (0.01-0.63)
0.017
0.09 (0.01-0.83)
0.033
ITCL + NK/T
Ref.
-
Ref.
-
Max tumor diameter
1.30 (1.13-1.48)
< 0.001
1.25 (1.07-1.47)
0.005
Invasive depth
Inside serous membrane
Ref.
-
Ref.
Serosal layer and outside
6.55 (2.67-16.06)
< 0.001
3.38 (1.24-9.19)
0.017
LDH
Elevated
0.67 (0.33-1.40)
0.287
Normal
Ref.
-
Virus
Yes
0.70 (0.23-2.14)
0.529
No
Ref.
-
Stage
I/II
Ref.
-
III/IV
0.82 (0.40-1.66)
0.581
ECOG
0-1
Ref.
-
Ref.
-
≥ 2
3.35 (1.02-11.05)
0.047
1.02 (0.24-4.39)
0.974
N-GCB
0.62 (0.29-1.31)
0.207
CD5 +
0.46 (0.17-1.25)
0.128
CD30 +
1.40 (0.42-4.68)
0.584
Double expression
Yes
0.40 (0.16-0.97)
0.044
0.51 (0.17-1.52)
0.227
No
Ref.
-
Ref.
-
Ki 67 ≥ 70%
Yes
1.13 (0.54-2.25)
0.747
No
Ref.
-
Macroscopic findings
Neoplasm
0.59 (0.12-2.88)
0.515
Ulceration
0.69 (0.33-1.42)
0.311
Others
Ref.
-
Imaging findings
Neoplasticlesions
0.49 (0.21-1.17)
0.109
Others
Ref.
-
Development of GICs risk predictive nomogram
Based on univariate and multivariate analyses, four parameters (primary tumor site, histopathological type, depth of tumor invasion, and max tumor diameter) were identified as independent risk factors. Hence, based on these four significant variables, a nomogram was created to predict risk of GICs in aggressive PGINHL patients (Figure 2). Add up the corresponding scores for each item and find the corresponding percentage on the GICs risk score.
Figure 2 Nomogram for predicting the risk of gastrointestinal complications in aggressive gastrointestinal non-Hodgkin lymphoma.
GICs: Gastrointestinal complications; ITCL: Intestinal T-cell lymphoma; NK: Nature killer; DLBCL: Diffuse large B-cell lymphoma.
Then, DCA showed that if the threshold probability was over 0.5, the nomogram for GICs prediction added more benefit than all or none, indicating that our nomogram provided a better clinical net benefit (Figure 3). As depicted in Figure 4, ROC analysis revealed that the nomogram exhibited favorable predictive performance for GICs. ROC curves were generated to further evaluate the predictive performance (area under curve = 0.815). Additionally, the calibration curves for the GICs probability exhibited optimal agreement (Figure 5).
Figure 4 The predictive performance of the risk nomogram for predicting gastrointestinal complications in aggressive gastrointestinal non-Hodgkin lymphoma.
Receiver operating characteristic curves displayed that this nomogram discriminated well. AUC: Area under the curve.
Figure 5 The calibration curve of risk nomogram of gastrointestinal complications in aggressive gastrointestinal non-Hodgkin lymphoma.
GICs: Gastrointestinal complications.
Treatment characteristics and prognosis of patients with GICs in aggressive PGINHL
Patients with aggressive PGINHL in the GICs group were treated differently from those in the No-GICs group (Table 3). In the former group, emergency surgery was performed because of GICs accounting for 88.3% in the GICs group, while in the No-GICs group, 3 patients underwent surgery because the tumors were not detected by endoscopy but were highly suspected by imaging. The rate of chemotherapy (71.7%) and radiotherapy (8.3%) in the GICs group were lower than those in the No-GICs group, probably because some patients could not tolerate chemotherapy because of poor ECOG for GICs, some patients improved after surgical resection of bulky masses and did not require radiotherapy. In conclusion, the patients with GICs were mainly given combination therapy (chemotherapy combined with surgery or radiotherapy). EFS and OS of GICs group were close to those in the No-GICs group (P > 0.05). At the end of follow-up, 26 patients (43.3%) died in the GIC group, and 20 patients (31.3%) died in the No-GIC group, there was no significant difference between the two groups (P > 0.05) (Figure 6).
Figure 6 Survival analysis of the gastrointestinal complications and No-gastrointestinal complications groups in primary gastrointestinal non-Hodgkin lymphomas.
A: Event-free survival curves for the comparison between gastrointestinal complications (GICs) and No-GICs group; B: Overall survival curves for the comparison between GICs and No-GICs group (P > 0.05). GICs: Gastrointestinal complications.
Table 3 Treatments and outcomes of patients in the No-gastrointestinal complications vs gastrointestinal complications groups of aggressive primary gastrointestinal non-Hodgkin lymphomas.
Characteristics
No-GICs (n = 64)
GICs (n = 60)
P value
Chemotherapy, n (%)
0.000
No
3 (4.7)
17 (28.3)
Yes
61 (95.3)
43 (71.7)
Radiotherapy, n (%)
0.022
No
49 (76.6)
55 (91.7)
Yes
15 (23.4)
5 (8.3)
Surgery, n (%)
0.000
No
61 (95.3)
5 (11.7)
Yes
3 (4.7)
53 (88.3)
Treatment, n (%)
Without CT
3 (4.7)
17 (28.3)
0.000
CT alone
43 (67.2)
4 (6.7)
CT + Sur/RT
18 (28.1)
39 (65.0)
EFS time, M (IQR)
21.0 (11.0, 36.0)
17.0 (5.5, 36.0)
0.580
OS time, M (IQR)
28.5 (14.5, 45.0)
24.5 (6.5, 47.5)
0.335
Death toll, n (%)
20 (31.3)
26 (43.3)
0.164
Analysis of prognostic factors in patients with aggressive PGINHL with GICs: Univariate analysis suggested that histopathology, LDH, stage, ECOG, non-GCB, chemotherapy combined with surgery or radiotherapy were related to OS (P < 0.05). Multivariate analysis suggested that histopathology, LDH, stage, ECOG, chemotherapy combined with surgery or radiotherapy were independent factors (P < 0.001, P = 0.005, P = 0.007, P < 0.001, P < 0.001 respectively) (Table 4).
Table 4 Univariate and multivariate analyses of factors associated with overall survival in aggressive primary gastrointestinal non-Hodgkin lymphoma patients with gastrointestinal complications.
Characteristics
Univariate analysis
Multivariate analysis
HR (95%CI)
P value
HR (95%CI)
P value
Age
≥ 60
1.29 (0.58-2.84)
0.535
< 60
Ref.
-
Gender
Male
1.83 (0.77-4.36)
0.174
Female
Ref.
-
Primary site
Esophagus or stomach
Ref.
-
Small intestine
1.53 (0.55-4.31)
0.417
Colon
1.25 (0.41-3.82)
0.697
Histopathology
ITCL + NKTCL
5.58 (2.40-13.00)
< 0.001
8.75 (2.87-26.71)
< 0.001
DLBCL + BL
Ref.
-
Ref.
-
Max tumor diameter
1.03 (0.93-1.14)
0.598
Invasive depth
Inside serous membrane
Ref.
-
Serosal layer and outside
1.15 (0.53-2.49)
0.728
LDH
Elevated
3.33 (1.51-7.33)
0.003
3.90 (1.50-10.12)
0.005
Normal
Ref.
-
Ref.
-
Virus
Yes
2.13 (0.73-6.23)
0.169
No
Ref.
-
Stage
I/II
Ref.
-
Ref.
-
III/IV
6.29 (2.35-16.82)
< 0.001
5.49 (1.58-19.09)
0.007
ECOG
0-1
Ref.
-
Ref.
-
≥ 2
13.40 (5.39-33.35)
< 0.001
24.14 (6.28-92.70)
< 0.001
N-GCB
0.28 (0.09-0.94)
0.040
CD5 +
1.47 (0.51-4.28)
0.479
CD30 +
1.58 (0.57-4.92)
0.347
Double expression
Yes
1.55 (0.58-4.12)
0.380
No
Ref.
-
KI 67 ≥ 70%
Yes
0.95 (0.42-2.12)
0.891
No
Ref.
-
Treatment
Without CT
Ref.
-
Ref.
-
CT alone
0.85 (0.24-2.97)
0.800
1.82 (0.44-7.50)
0.407
CT + Sur/RT
0.07 (0.03-0.19)
< 0.001
0.07 (0.02-0.24)
< 0.001
DISCUSSION
Primary gastrointestinal lymphoma can occur in any part of the digestive tract, most commonly in the stomach, followed by the small intestine, colon and rectum, rarely in the esophagus[25,26]. The most common pathological type of PGINHL was DLBCL, followed by MALT, MCL, rare types included ITCL, BL, NKTCL, etc.[27]. More attention should be paid to aggressive gastrointestinal lymphomas such as DLBCL, BL, ITCL, and NKTCL. Owing to the rapid progression of the aggressive lymphomas, the enlargement and invasive characteristics of the mass were easy to cause GICs, the management of GICs remained a clinical challenge[22]. In our study of aggressive gastrointestinal lymphoma cohort, GICs occurred in 48.4% of patients. The most common histological type in GICs group was DLBCL (81.7%). The most common anatomical site in the GICs group was the small intestine (43.3%). Some studies suggested that primary small intestinal DLBCL was indeed more likely to develop intestinal obstruction or perforation[28]. In our GICs group, bulky lymphoma accounted for a relatively high proportion, the maximum tumor diameter in this group was approximately 7.03 cm ± 3.7 cm, which was also a factor that leaded to more GIO in this group. In our study, GICs occurred in 11 cases (91.7%) of 12 patients with PGI-TNKL, while 49 patients (43.8%) of the 112 patients with PGI-B cell lymphoma developed GICs, which including 36 cases of GIO, 9 cases of GIP, 12 cases of GIB, and 3 cases of GIP and GIB. ITCL was more prone to perforation than intestinal B-cell lymphoma[14,29].
Multivariate Logistic regression analysis showed that the four independent risk factors for GICs were the primary site of small intestine, pathological type PGI-NKTL, large tumor diameter, invasion of the serosal layer and beyond. First, regarding the discussion of the most common primary site, a study suggested that the small intestine was the most common site, accounting for 59%[22]. In the single-center study on intestinal lymphoma complicated with perforation, 55% (51/92) of the cases occurred after chemotherapy[30]. Most patients had GIO before chemotherapy, accounting for 60% (36/60), and there were 20% (12/60) cases with intestinal perforation before or after chemotherapy. In our study, the highest occurrence site of GICs was also in the small intestine, accounting for 43.3%. The decrease in perforation rate in our study may be related to our early surgical intervention. Second, regarding the most common histopathological types, a study suggested that aggressive B-cell lymphoma had a higher risk of perforation than indolent B-cell lymphoma, particularly DLBCL[22]. In addition, although the incidence of enteric T-cell lymphoma was lower, GIP was higher in aggressive T-cell lymphoma[31]. In our study, the most common histological type of GICs group was DLBCL (81.7%), followed by ITCL (16.7%), and NKTCL (1.6%). However, GICs occurred in 11 of the 12 patients with PGI-TNKL, especially intestinal perforation, required emergency surgery. At the end of our follow-up, only one patient in PGI-TNKL survived, and the others died within a year.
Third, the depth of tumor invasion of the tube wall was related to the occurrence of GICs. Studies have suggested that patients with gastrointestinal cancer were prone to perforation once the tumor encroached on the serosal layer[32]. Microscopic findings showed that alterations in gastrointestinal lymphoma began in the submucosal layer where they were most extensive, rarely spread to the muscular layer, and serosal layer[33]. Fourth, regarding the correlation between the maximum tumor diameter and the risk of GICs, the main reason was digestive tract obstruction caused by bulky mass. In a retrospective study of DLBCL, patients with tumor mass ≥ 10 cm, or intestinal involvement had significantly higher risk of severe GICs as initial manifestations[7].
Summarizing our study and previous clinical studies, these four factors did have GICs correlation, and we developed a nomogram to predict risk of GICs in aggressive PGINHL patients based on independent risk factors. The higher the total scores of the corresponding nomogram, the greater the risk of GICs. The high risk of GICs suggested that clinicians needed to take emergency measures such as surgical intervention and postponing chemotherapy. DCA mapping showed that our predictive model provided better clinical benefits. The ROC and calibration curves for the GICs probability exhibited an optimal agreement. Currently, there were no predictive GICs risk models for aggressive intestinal lymphoma. A small sample study proposed a risk scoring system that included indolent PGI lymphoma patients to predict severe GICs events, patients with high scores were candidates for elective surgery to prevent GIC events[34]. Our predictive GICs risk nomogram covered a more susceptible population than the few previously reported, with a relatively many predictors. Regarding prognostic markers for large B-cell lymphoma (Double expression, CD5, no-GCB), these markers did not hold up in multivariate analyses in this study population and it may be worthwhile to look specifically at their role in patients with large B-cell lymphoma. The limitation of this study was that there was a certain bias in the study data because it was retrospective. Also, it was the absence of external validation. In the future, multi-center and prospective studies are needed to verify and improve these results.
Unlike the patients with nodal or other primary extra nodal lymphomas, PGINHL patients with GICs required immediate intervention. because GICs were life-threatening acute abdomens such as GIO, GIB, and GIP. In particular, the perforation rate of small intestine was more than 50%, and the mortality rate was high[30]. How to adjust the treatments of those patients? Priority surgery, delayed surgery, or reduced-dose chemotherapy? In the GICs cohort we studied emergency surgery was performed, accounting for 88.3%. In the No-GICs group, three patients underwent surgery because the tumors were not detected by endoscopy but highly suspected by imaging. The majority of patients with GICs were given combination therapy. Due to identification of emergency and timely treatment of GICs, local surgery can prevent spontaneous perforation and/or bleeding during chemotherapy or radiotherapy and provide opportunities for subsequent systemic treatment. Patients with good recovery after total resection were treated with immunochemotherapy. Adjuvant radiotherapy was given to bulky patients who could not be totally resected. For inoperable patients, attention should be paid to the initial chemotherapy starting from a low dose and gradually increasing to a standard dose. The tumor regression should be closely observed. Lymphoma was sensitive to immunochemotherapy and retreated rapidly; and normal tissue could not be repaired because of bulky or deep invasion lesions in the gastrointestinal tract. Thus, GICs caused by tumor regression after high-dose chemotherapy should be avoided as much as possible. Even if they undergo emergency surgery, faced severe myelosuppression after chemotherapy and had a poor prognosis. Therefore, we need to summarize the characteristics of patients with aggressive PGINHL and expand the sample size to avoid small sample bias.
Our survival analysis in GICs group, suggested that histopathology, LDH, stage, ECOG, chemotherapy combined with surgery or radiotherapy were independent factor. It had been reported that high LDH, advanced stage, and high ECOG in the international prognostic index scoring system were associated with poor OS of NHL[35,36]. In our study, ten cases of PGI-TNKL died of intestinal perforation and bleeding, the prognosis were poor. A multi-center prospective study in Asia suggested that PGI-TNKL showed aggressive behavior and poor OS[37]. Aggressive clinicopathological features of PGI-TNKL were different from indolent T-cell lymphoproliferative disorder[4]. Microscopic examination showed a diffuse transmural lymphoid infiltrate with accompanying mucosal ulceration[38]. Therefore, both clinicians and pathologists must be aware of the distinct characteristics of these lesions to ensure that appropriate care was provided. In our study, combination therapy did improve survival of the patients with GICs, EFS and OS were close to those in the No-GICs group. However, Surgery was controversial in GI lymphoma[39-41], we should grasp the surgical indications and avoid meaningless invasive surgery in low-risk populations.
CONCLUSION
The complication rate of GICs in patients with aggressive PGINHL was relatively high, especially in those with PGI-DLBCL. The independent risk factors for GICs were the small intestine, PGI-TNKL, bulky tumor, and depth of invasion. A preliminary risk model for predicting GICs was established. We analyzed therapeutic interventions for patients with a high incidence of GICs. Comprehensive treatment including surgery improved the OS of patients with GICs. We hope to further validate the nomogram with larger samples in the future.
Footnotes
Provenance and peer review: Unsolicited article; Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Oncology
Country of origin: China
Peer-review report’s classification
Scientific Quality: Grade B
Novelty: Grade B
Creativity or Innovation: Grade B
Scientific Significance: Grade B
P-Reviewer: Frolli A S-Editor: Fan M L-Editor: A P-Editor: Zhao S
Mansoor A, Pittaluga S, Beck PL, Wilson WH, Ferry JA, Jaffe ES. NK-cell enteropathy: a benign NK-cell lymphoproliferative disease mimicking intestinal lymphoma: clinicopathologic features and follow-up in a unique case series.Blood. 2011;117:1447-1452.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 115][Cited by in F6Publishing: 122][Article Influence: 8.1][Reference Citation Analysis (0)]
Kohri M, Tsukasaki K, Akuzawa Y, Tanae K, Takahashi N, Saeki T, Okamura D, Ishikawa M, Maeda T, Kawai N, Matsuda A, Arai E, Arai S, Asou N. Peripheral T-cell lymphoma with gastrointestinal involvement and indolent T-lymphoproliferative disorders of the gastrointestinal tract.Leuk Res. 2020;91:106336.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 11][Cited by in F6Publishing: 11][Article Influence: 2.2][Reference Citation Analysis (0)]
Kim SJ, Choi CW, Mun YC, Oh SY, Kang HJ, Lee SI, Won JH, Kim MK, Kwon JH, Kim JS, Kwak JY, Kwon JM, Hwang IG, Kim HJ, Lee JH, Oh S, Park KW, Suh C, Kim WS. Multicenter retrospective analysis of 581 patients with primary intestinal non-hodgkin lymphoma from the Consortium for Improving Survival of Lymphoma (CISL).BMC Cancer. 2011;11:321.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 81][Cited by in F6Publishing: 86][Article Influence: 6.1][Reference Citation Analysis (0)]
Kim DH, Lee D, Kim JW, Huh J, Park SH, Ha HK, Suh C, Yoon SM, Kim KJ, Choi KD, Ye BD, Byeon JS, Song HJ, Jung HY, Yang SK, Kim JH, Myung SJ. Endoscopic and clinical analysis of primary T-cell lymphoma of the gastrointestinal tract according to pathological subtype.J Gastroenterol Hepatol. 2014;29:934-943.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 25][Cited by in F6Publishing: 30][Article Influence: 2.7][Reference Citation Analysis (0)]
Cheson BD, Fisher RI, Barrington SF, Cavalli F, Schwartz LH, Zucca E, Lister TA; Alliance, Australasian Leukaemia and Lymphoma Group; Eastern Cooperative Oncology Group; European Mantle Cell Lymphoma Consortium; Italian Lymphoma Foundation; European Organisation for Research; Treatment of Cancer/Dutch Hemato-Oncology Group; Grupo Español de Médula Ósea; German High-Grade Lymphoma Study Group; German Hodgkin's Study Group; Japanese Lymphorra Study Group; Lymphoma Study Association; NCIC Clinical Trials Group; Nordic Lymphoma Study Group; Southwest Oncology Group; United Kingdom National Cancer Research Institute. Recommendations for initial evaluation, staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: the Lugano classification.J Clin Oncol. 2014;32:3059-3068.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 2850][Cited by in F6Publishing: 3600][Article Influence: 327.3][Reference Citation Analysis (0)]
Choi WW, Weisenburger DD, Greiner TC, Piris MA, Banham AH, Delabie J, Braziel RM, Geng H, Iqbal J, Lenz G, Vose JM, Hans CP, Fu K, Smith LM, Li M, Liu Z, Gascoyne RD, Rosenwald A, Ott G, Rimsza LM, Campo E, Jaffe ES, Jaye DL, Staudt LM, Chan WC. A new immunostain algorithm classifies diffuse large B-cell lymphoma into molecular subtypes with high accuracy.Clin Cancer Res. 2009;15:5494-5502.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 488][Cited by in F6Publishing: 464][Article Influence: 29.0][Reference Citation Analysis (0)]
Nakamura S, Matsumoto T, Takeshita M, Kurahara K, Yao T, Tsuneyoshi M, Iida M, Fujishima M. A clinicopathologic study of primary small intestine lymphoma: prognostic significance of mucosa-associated lymphoid tissue-derived lymphoma.Cancer. 2000;88:286-294.
[PubMed] [DOI][Cited in This Article: ][Cited by in F6Publishing: 4][Reference Citation Analysis (0)]
Moayyedi P, Eikelboom JW, Bosch J, Connolly SJ, Dyal L, Shestakovska O, Leong D, Anand SS, Störk S, Branch KRH, Bhatt DL, Verhamme PB, O'Donnell M, Maggioni AP, Lonn EM, Piegas LS, Ertl G, Keltai M, Cook Bruns N, Muehlhofer E, Dagenais GR, Kim JH, Hori M, Steg PG, Hart RG, Diaz R, Alings M, Widimsky P, Avezum A, Probstfield J, Zhu J, Liang Y, Lopez-Jaramillo P, Kakkar A, Parkhomenko AN, Ryden L, Pogosova N, Dans A, Lanas F, Commerford PJ, Torp-Pedersen C, Guzik T, Vinereanu D, Tonkin AM, Lewis BS, Felix C, Yusoff K, Metsarinne K, Fox KAA, Yusuf S; COMPASS Investigators. Pantoprazole to Prevent Gastroduodenal Events in Patients Receiving Rivaroxaban and/or Aspirin in a Randomized, Double-Blind, Placebo-Controlled Trial.Gastroenterology. 2019;157:403-412.e5.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 87][Cited by in F6Publishing: 104][Article Influence: 17.3][Reference Citation Analysis (0)]
Koch P, del Valle F, Berdel WE, Willich NA, Reers B, Hiddemann W, Grothaus-Pinke B, Reinartz G, Brockmann J, Temmesfeld A, Schmitz R, Rübe C, Probst A, Jaenke G, Bodenstein H, Junker A, Pott C, Schultze J, Heinecke A, Parwaresch R, Tiemann M; German Multicenter Study Group. Primary gastrointestinal non-Hodgkin's lymphoma: I. Anatomic and histologic distribution, clinical features, and survival data of 371 patients registered in the German Multicenter Study GIT NHL 01/92.J Clin Oncol. 2001;19:3861-3873.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 326][Cited by in F6Publishing: 265][Article Influence: 11.0][Reference Citation Analysis (0)]
Maeshima AM, Taniguchi H, Ito Y, Hatta S, Suzuki T, Yuda S, Makita S, Fukuhara S, Munakata W, Suzuki T, Maruyama D, Izutsu K. Clinicopathological characteristics of diffuse large B-cell lymphoma involving small and large intestines: an analysis of 126 patients.Int J Hematol. 2019;110:340-346.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 6][Cited by in F6Publishing: 11][Article Influence: 1.8][Reference Citation Analysis (0)]
Daum S, Ullrich R, Heise W, Dederke B, Foss HD, Stein H, Thiel E, Zeitz M, Riecken EO. Intestinal non-Hodgkin's lymphoma: a multicenter prospective clinical study from the German Study Group on Intestinal non-Hodgkin's Lymphoma.J Clin Oncol. 2003;21:2740-2746.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 186][Cited by in F6Publishing: 162][Article Influence: 7.4][Reference Citation Analysis (0)]
Modemann F, Ahmadi P, von Kroge PH, Weidemann S, Bokemeyer C, Dierlamm J, Fiedler W, Ghandili S. The prognostic impact of lymphoma perforation in patients with primary gastrointestinal lymphoma - a single-center analysis.Leuk Lymphoma. 2023;64:1801-1810.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Wang RC, Chen BJ, Yuan CT, Ho CH, Chuang WY, Chen SW, Chang JH, Yu WH, Chuang SS. The spectrum of intestinal mature T- and NK-cell neoplasms in a tertiary center in Taiwan with a high frequency of perforation.Pathol Res Pract. 2022;240:154184.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Yoon SE, Song Y, Kim SJ, Yoon DH, Chen TY, Koh Y, Kang KW, Lee HS, Tay KKW, Lim ST, Poon M, Irawan C, Zhao W, Do YR, Lee MH, Ng SC, Lee WS, Guo Y, Zhang H, Kang HJ, Yun HJ, Kim HJ, Lung DTC, Kwak JY, Han JJ, Mun YC, Oh SY, Shim H, Kwon JH, Sohn BS, Park SK, Jo JC, Ko YH, Jun Z, Kim WS. Comprehensive analysis of peripheral T-cell and natural killer/T-cell lymphoma in Asian patients: A multinational, multicenter, prospective registry study in Asia.Lancet Reg Health West Pac. 2021;10:100126.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 8][Cited by in F6Publishing: 34][Article Influence: 8.5][Reference Citation Analysis (0)]
Tan SY, Chuang SS, Tang T, Tan L, Ko YH, Chuah KL, Ng SB, Chng WJ, Gatter K, Loong F, Liu YH, Hosking P, Cheah PL, Teh BT, Tay K, Koh M, Lim ST. Type II EATL (epitheliotropic intestinal T-cell lymphoma): a neoplasm of intra-epithelial T-cells with predominant CD8αα phenotype.Leukemia. 2013;27:1688-1696.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 91][Cited by in F6Publishing: 97][Article Influence: 8.1][Reference Citation Analysis (0)]
Mehmet K, Sener C, Uyeturk U, Seker M, Tastekin D, Tonyali O, Balakan O, Yazici OK, Urakci Z, Isikdogan A, Ozdemir N, Inal A, Kaplan MA, Suner A, Dal S, Uncu D, Gumus M, Boruban MC, Oksuzoglu B, Ayyildiz O, Benekli M. Treatment modalities in primary gastric lymphoma: the effect of rituximab and surgical treatment. A study by the Anatolian Society of Medical Oncology.Contemp Oncol (Pozn). 2014;18:273-278.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 1][Cited by in F6Publishing: 6][Article Influence: 0.5][Reference Citation Analysis (0)]
