Integrated Chinese and Western medicine prolong survival in colorectal cancer with liver metastasis: A retrospective cohort study

Abstract

BACKGROUND

Colorectal cancer is a leading malignancy worldwide, with the liver being the most common site of metastasis and the primary cause of death. While chemotherapy improves overall survival (OS) in colorectal cancer liver metastases (CRLM), conventional Western medicine has limited efficacy. In the previous cohort study on CRLM conducted by our team, it was observed that integrated Chinese and Western medicine (ICWM) may offer advantages in treating CRLM, but high-level evidence is still lacking.

AIM

To analyze the survival benefits of ICWM treatment in patients with CRLM.

METHODS

A retrospective cohort study was conducted, with “continuous traditional Chinese medicine (TCM) treatment for ≥ 3 months” as the exposure factor. A total of 499 patients from four tertiary grade A hospitals were divided into two cohorts: The ICWM treatment cohort (cohort A) with 185 patients and the Western medicine treatment cohort (cohort B) with 314 patients. OS, 3-year/5-year survival rates, and subgroup efficacy were compared. Multivariate Cox proportional hazards regression was employed to identify independent prognostic factors.

RESULTS

The study demonstrated a significant improvement in OS for patients receiving ICWM treatment (cohort A) compared to Western medicine alone (cohort B). The median OS was 36 months vs 28.37 months (hazard ratio = 0.65, P < 0.01), with higher 3-year and 5-year survival rates (47.03% vs 35.99%; 19.46% vs 10.51%). Subgroup analysis revealed statistically significant differences (P < 0.05) between two cohorts in subgroups such as age, sex, the primary tumor site, different genotypes, whether local treatment of liver metastases, treatment stage and whether there is concurrent extrahepatic metastasis were administered. Multivariate analysis identified right-sided colon location and rat sarcoma viral oncogene homolog/B-Raf proto-oncogene, serine/threonine kinase mutations as poor prognostic factors, while local treatment of liver metastases and integrated TCM therapy were protective, reducing mortality risk by 39% and 47%, respectively.

CONCLUSION

ICWM treatment is associated with longer overall survival in CRLM patients and improve their 3^rd and 5^th survival rates, with TCM integration serving as an independent protective factor.

Key Words: Colorectal cancer liver metastases; Integration of Chinese and Western medicine; Cohort study; Local treatment; Overall survival

Core Tip: The liver is the most common site of metastasis for colorectal cancer and the primary cause of death among patients with this disease. This retrospective cohort study confirms that in real-world settings, integrated Chinese and Western medicine treatment is associated with longer overall survival in patients with liver metastases from colorectal cancer. Additionally, for patients unsuitable for local treatment, integrated Chinese and Western medicine therapy has also been shown to correlate with prolonged survival.

INTRODUCTION

Epidemiological analyses to date indicate that colorectal cancer (CRC) ranks as the second most prevalent malignancy among women and the third most prevalent among men worldwide. The liver is recognized as the most common site of metastasis, and hepatic metastases remain the primary cause of mortality in CRC patients[1]. Previous studies have reported that the overall survival (OS) of untreated patients with CRC liver metastases (CRLM) ranges from 4.5 months to 12 months, whereas chemotherapy can extend OS to 15-20 months[2]. Currently, conventional Western medicine faces therapeutic limitations that restrict improvements in the OS of CRC patients with liver metastases[3]. In a previous cohort study on the effect of integrated Chinese and Western medicine (ICWM) treatment on the survival of patients with metastatic CRC, our team showed that traditional Chinese medicine (TCM) has unique advantages in the treatment of CRC with liver metastasis[4]. However, the efficacy of this treatment lacks high-level evidence-based medical evidence and is not yet widely recognized. Therefore, this study aims to analyze the survival benefits of ICWM treatment in patients with CRLM.

MATERIALS AND METHODS

Study design

A retrospective cohort study was conducted, including patients with CRLM who were admitted to Xiyuan Hospital of the China Academy of Chinese Medical Sciences, Xinjiang Uygur Autonomous Region Hospital of TCM, Dongfang Hospital of Beijing University of Chinese Medicine, and Guang’anmen Hospital of the China Academy of Chinese Medical Sciences between August 1, 2013 and December 31, 2022. All enrolled patients received standard Western medical therapy, including but not limited to chemotherapy and/or targeted therapy. Previous research by our team demonstrated that receiving integrated TCM therapy for more than three months significantly improves prognosis in CRC patients[4]. Accordingly, continuous TCM treatment for ≥ 3 months was considered the exposure factor. Those who meet this exposure factor are included in the ICWM treatment cohort (cohort A) and receive TCM in combination with standard Western medicine treatment; those who do not meet this exposure factor are included in the Western medicine treatment cohort (cohort B) and receive standard Western medicine treatment. Follow-up was conducted to compare whether there were differences in survival benefits between the two patient cohorts. This study has been approved by the Ethics Review Committee of the Xiyuan Hospital of the China Academy of Chinese Medical Sciences (Approval No. 2025XLA052-1) and registered at the International Traditional Medicine Clinical Trial Registry (No. ITMCTR2025001663).

Inclusion criteria: (1) Age ≥ 18 years; (2) Pathologically diagnosed with CRLM, clinical stage IV; (3) Treatment stage unrestricted; (4) Electrocorticography performance status 0-2; (5) Expected survival of ≥ 3 months; (6) Measurable lesions meeting Response Evaluation Criteria in Solid Tumors 1.1 criteria; (7) No severe abnormalities in cardiac, pulmonary, hepatic, or renal function; (8) Normal coagulation function, no active bleeding or thrombotic disorders; and (9) Good compliance expected, able to follow up on treatment efficacy and adverse reactions as per the protocol.

Exclusion criteria: (1) Patients with clinically symptomatic brain metastases; (2) Patients with concurrent unresolved malignant tumors; (3) Patients confirmed by immunohistochemistry to have defective mismatch repair or by genetic testing to have high microsatellite instability; (4) Patients whose overall condition is unsuitable for conventional anticancer therapy; (5) Patients who are about to participate in or are currently participating in other clinical trials; (6) Patients with other conditions deemed unsuitable for inclusion by the investigator, such as pregnancy; and (7) Patients whose medical records indicate refusal to use their cases for research purposes.

Dropout criteria

Patients who cannot be followed up in the outpatient clinic and cannot be reached after three telephone follow-ups (no answer, phone turned off, refusal to answer) or whose contact number is invalid are considered lost to follow-up (dropout) cases. Outcome measures are calculated up to the last follow-up date.

Intervention measures

Western medicine cohort (cohort B): Receive standard Western medicine treatment, including chemotherapy and/or targeted therapy recommended in the National Comprehensive Cancer Network, Chinese Society of Clinical Oncology, and European Society for Medical Oncology CRC guidelines, combined with local treatment measures such as radiotherapy, surgery, intervention, and radiofrequency ablation.

ICWM cohort (cohort A): In addition to standard Western medicine treatment, patients receive continuous TCM treatment for ≥ 3 months. The therapeutic principles of TCM are defined as follows: (1) First-line and second-line induction chemotherapy phase: Primarily focus on strengthening the body’s vital energy; and (2) Maintenance therapy, third-line and subsequent treatment phases: Simultaneously strengthen the body’s vital energy and eliminate pathogenic factors. Therapeutic principles for eliminating pathogenic factors include “clearing heat and detoxifying”, “softening hardness and dispersing nodules”, and “promoting blood circulation and resolving stasis”. Both decoctions and proprietary Chinese medicines may be used. “TCM treatment” is defined as: (1) Continuous administration of TCM decoctions based on the theory of combining disease diagnosis and syndrome differentiation, incorporating the principle of “tonifying the vital Qi and dispelling pathogenic factors”; and (2) Under the guidance of a TCM physician, the use of anticancer TCM formulations with effects such as “clearing heat and detoxifying”, “softening hardness and dispersing nodules”, and “promoting blood circulation and resolving stasis”, including but not limited to Antike Capsules, Pingxiao Capsules, Compound Banmao Capsules, Xihuang Capsules (Pills), and Weimaining Capsules. Any of the above scenarios are defined as receiving “TCM treatment”. Non-pharmacological treatments such as acupuncture, massage, and qigong are not considered “TCM treatment”. “Continuous TCM treatment” is defined as receiving TCM treatment for ≥ 5 days per week. All patients were admitted to tertiary-level hospitals, and Western medical treatments were conducted in accordance with guidelines and diagnostic and treatment protocols to ensure that all enrolled patients received the best treatment available at the time.

Follow-up

Follow-up will be conducted via outpatient visits or telephone calls. In addition to irregular outpatient visits, telephone follow-ups will be conducted every three months. Follow-up will continue until May 30, 2025, or until the patient’s death. The focus of follow-up will be on the patient’s survival status, whether they have received continuous TCM treatment for more than three months, and any major adverse reactions during treatment.

Observation indicators

OS is calculated from the time of initial diagnosis of liver metastasis to the time of death. For patients who have not died, OS is calculated from the enrollment date to May 30, 2025, and is set as censored. For patients who are lost to follow-up, OS is calculated from the enrollment date to the last follow-up date before loss to follow-up and is set as censored. Three- and five-year cumulative survival rates are calculated.

Outcome measures

Primary outcome measure: Difference in OS between the two cohorts.

Secondary outcome measure: Cumulative survival rates at 3 years and 5 years for the two cohorts.

Statistical analysis

Data collection was performed using Excel, and statistical analysis was conducted using SAS JMP Pro 14.0. Continuous variables that followed a normal distribution were expressed as mean ± SD, while those that did not follow a normal distribution were expressed as median (interquartile rage). Baseline analysis used χ² tests to assess balance between the two cohorts, including age, gender, primary tumor site, whether local treatment was performed for liver metastases, gene type, and metastasis site. For samples insufficient for χ², the Fisher’s exact probability test was used. For the primary endpoint OS, survival curves were plotted using the Kaplan-Meier method, and intergroup comparisons were performed using the log-rank test to calculate 3-year and 5-year survival rates. Intergroup comparisons were performed using χ². The significance level was set at α = 0.05.

RESULTS

General situation

A total of 194 patients were screened in the cohort A, with 185 enrolled; a total of 325 patients were screened in the cohort B, with 314 enrolled. As of the last follow-up date (May 30, 2025), 139 patients in the cohort A reached the clinical endpoint (death), with 8 lost to follow-up. In the cohort B, 269 patients reached the clinical endpoint (death), and 4 were lost to follow-up. The reasons for loss to follow-up in both cohorts were three consecutive unsuccessful phone calls or invalid phone numbers. Twelve patients initially received only Western medicine treatment but were later found to have received continuous TCM treatment for over three months during follow-up, and were therefore transferred to the cohort A. Demographic and baseline data comparisons between the two cohorts are shown in Table 1.

Table 1 Comparison of baseline data between two cohorts, mean ± SD/n (%).

	Cohort A (n = 185)	Cohort B (n = 314)	t/χ2	P value
Age	62.47 ± 11.30	58.51 ± 10.33	3.89	< 0.01a
Gender			0.60	0.44
Male	106 (57.30)	191 (60.83)
Female	79 (42.70)	123 (39.17)
Primary site			0.08	0.78
Right-side colon	57 (30.81)	93 (29.62)
Left-side colon	128 (69.19)	221 (70.38)
Local treatment of liver metastases1			40.60	< 0.01a
Yes	66 (35.68)	204 (64.97)
No	119 (64.32)	110 (35.03)
Gene type			0.81	0.67
RAS/BRAF wild type	83 (44.86)	141 (44.90)
RAS/BRAF mutation	87 (47.03)	154 (49.04)
No genetic testing	15 (8.11)	19 (6.05)
Site of metastasis			0.13	0.72
Liver metastasis only	73 (39.46)	129 (41.08)
Combined with extrahepatic metastasis	112 (60.54)	185 (58.92)

^aP < 0.01.

¹Local treatments include radiofrequency ablation, local interventional surgical resection, radiotherapy, and argon-helium knife therapy.

RAS: Rat sarcoma viral oncogene homolog; BRAF: B-Raf proto-oncogene, serine/threonine kinase.

Outcome measures

OS: The median OS (mOS) for patients with CRLM in the cohort A was 36 months [95% confidence interval (CI): 32-43], while the mOS for the cohort B was 28.37 months (95%CI: 25.6-31.43). The difference was statistically significant [hazard ratio (HR) = 0.65, P < 0.01; Figure 1].

Figure 1 Overall survival. Blue curve: Cohort A; red curve: Cohort B.

Cumulative survival rates at 3 years and 5 years: The survival rates at 3 years and 5 years for the two cohorts are shown in Table 2.

Table 2 Cumulative survival rates at 3 years and 5 years.

	Cohort A	Cohort B	χ2	P value
3-year survival rate	47.03%	35.99%	5.88	0.02a
5-year survival rate	19.46%	10.51%	7.58	0.01a

^aP < 0.05.

Subgroup analysis: In subgroups stratified by age, gender, primary tumor site, different genotypes [rat sarcoma viral oncogene homolog (RAS)/B-Raf proto-oncogene, serine/threonine kinase (BRAF) wild-type/mutant], receipt of local treatment for liver metastases, treatment stage, and presence of extrahepatic metastases, the OS differences between the two cohorts were statistically significant (Table 3 and Figure 2). Among the 34 patients without genetic testing (15 in the cohort A and 19 in the cohort B), only patients with confirmed genetic testing results were analyzed in the following genetic subtype subgroups.

Figure 2 Survival curve. A: Non-elderly subgroup (< 65 years); B: Elderly subgroup(≥ 65 years); C: Survival curve of male subgroup; D: Female subgroup; E: The left-side colon subgroup; F: The right-side colon subgroup; G: The rat sarcoma viral oncogene homolog/B-Raf proto-oncogene, serine/threonine kinase wild-type subgroup; H: The rat sarcoma viral oncogene homolog/B-Raf proto-oncogene, serine/threonine kinase mutation subgroup; I: The liver metastasis local treatment subgroup; J: The non-liver metastasis local treatment subgroup; K: Before the third line treatment subgroup; L: After the third line treatment subgroup; M: The only liver metastasis subgroup; N: The combined with extrahepatic metastasis subgroup. Blue curve: Cohort A; red curve: Cohort B.

Table 3 Subgroup analysis forest plot.

Subgroup	mOS/month (95%CI)		χ2	HR (95%CI)	P value
	Cohort A	Cohort B
Age
< 65 years	39.87 (33-48.5)	28.62 (25.4-31.8)	15.16	0.59 (0.45-0.77)	< 0.01b
≥ 65 years	31 (28-41)	27.13 (22.3-32.47)	4.58	0.70 (0.50-0.97)	0.03a
Sex
Male	36 (31-45.27)	30.57 (26.4-34.53)	6.25	0.70 (0.53-0.92)	0.01a
Female	35.57 (30.8-44)	25.37 (22.53-28.4)	12.55	0.70 (0.51-0.97)	< 0.01b
Side
Left	36.53 (32.47-44)	32.43 (28.73-36.23)	6.50	0.73 (0.57-0.93)	0.01a
Right	32 (28.4-44)	19.47 (15.23-23.3)	13.25	0.50 (0.34-0.73)	< 0.01b
Gene
RAS/BRAF wild-type	41.57 (35-57.5)	32.6 (27.87-36.07)	9.22	0.62 (0.45-0.84)	< 0.01b
RAS/BRAF mutant	32.47 (28.4-40.6)	26.38 (22.53-29.63)	6.97	0.67 (0.50-0.90)	< 0.01b
Local treatment of liver metastases
Yes	48.43 (35.5-64)	31.25 (28.27-34.43)	13.25	0.54 (0.38-0.75)	< 0.01b
No	31.9 (28.8-39)	20.28 (17.2-24.37)	13.60	0.59 (0.45-0.79)	< 0.01b
Treatment lines
Before the third line	32.7 (30-41)	25.8 (23.3-28.93)	9.72	0.69 (0.55-0.87)	< 0.01b
After the third line	43.73 (35-57.5)	32.6 (29.8-36.7)	11.14	0.48 (0.30-0.74)	< 0.01b
Only liver metastasis	33.53 (28.4-62.9)	30.7 (27.4-36.23)	6.60	0.65 (0.46-0.90)	0.01a
Combined with extrahepatic metastasis	36.53 (31.4-43)	25.8 (21.3-29.63)	10.82	0.65 (0.50-0.84)	< 0.01b

^aP < 0.05.

^bP < 0.01.

RAS: Rat sarcoma viral oncogene homolog; BRAF: B-Raf proto-oncogene, serine/threonine kinase; mOS: Median overall survival; HR: Hazard ratio; CI: Confidence interval.

Cox multivariate analysis of the entire population

After merging the data from the cohort A and the cohort B, a Cox multivariate risk ratio regression model was used to explore factors influencing survival in CRLM patients. The following independent variables were included: Primary site (right-colon/Left-colon), gene status (RAS/BRAF wild-type, RAS/BRAF mutant-type), local treatment for liver metastases, and cohort (cohort A/cohort B). The results showed that right colon and RAS/BRAF gene mutations were independent factors associated with poor prognosis, while local treatment of liver metastases and combined TCM treatment were protective factors, reducing the risk of death by 39% and 47%, respectively (Table 4).

Table 4 Cox multifactor analysis.

Factor	SEM	χ2	ν	P value	HR (95%CI)
Primary site (right/Left-side colon)	0.056	7.769	1	0. 006b	1.364 (1.093-1.692)
Gene type (RAS/BRAF wild type RAS/BRAF mutation)	0.086	8.490	2	0.022a	0.785 (0.639-0.965)
Local treatment of liver metastases (yes/no)	0.053	21.197	1	< 0.001b	0.616 (0.501-0.757)
Cohort (ICWM/WM)	0.056	31.773	1	< 0.001b	0.533 (0.427-0.662)

^aP < 0.05.

^bP < 0.01.

RAS: Rat sarcoma viral oncogene homolog; BRAF: B-Raf proto-oncogene, serine/threonine kinase; ICWM: Integrated Chinese and Western medicine; WM: Western medicine; HR: Hazard ratio; CI: Confidence interval.

DISCUSSION

This study is the first to use a retrospective cohort study method to confirm that, compared with standard Western medicine treatment alone, ICWM treatment can effectively improve the clinical efficacy of CRLM patients and is associated with longer OS, providing evidence-based medical evidence for the ICWM treatment of CRLM.

Baseline analysis showed that only 35.68% of patients in the cohort A underwent local treatment for liver metastases, while most patients (64.97%) in the cohort B underwent local treatment for liver metastases. At the same time, the average age of patients in the cohort A was significantly higher than that of patients in the cohort B, with a high proportion of elderly patients and patients who were treated at a later stage, resulting in a generally poor overall condition. Survival analysis showed that the mOS for the cohort A and cohort B was 36 months and 28.37 months, respectively, with a statistically significant difference (HR = 0.65, P < 0.01). Integrated TCM treatment reduced the risk of death from CRLM by 35%. The survival rates at 3 years and 5 years in the cohort A were 11.04% and 8.59% higher than those in the cohort B, respectively, with statistically significant differences. This indicates that combined TCM treatment is associated with significantly longer OS in CRLM patients compared to standard Western medicine treatment alone. Subgroup analysis revealed that the standard first-line to third-line treatments for metastatic CRC typically follow relatively uniform regimens, such as capecitabine and oxaliplatin, folinic acid/fluorouracil/oxaliplatin/, capecitabine and irinotecan, folinic acid/fluorouracil/oxaliplatin/irinotecan, and fruquintinib/regorafenib/trifluridine and tipiracil, etc.; while the diversity and personalization of treatment regimens after the third line have improved, patients’ progression-free survival (PFS) is typically shorter than in earlier lines. The results of this study showed that the mOS of patients in the cohort A and cohort B before and after the third-line treatment stage were 32.7 vs 25.8 and 43.73 vs 32.6, respectively. Combined TCM treatment reduced the risk of death by 31% and 52%, respectively. These results further validate that TCM treatment can significantly prolong OS in both the first-line and subsequent treatment stages. Right-sided colon cancer and RAS/BRAF gene mutations are widely recognized as poor prognostic factors for CRLM[5], while ICWM therapy is a protective factor, reducing the risk of death by 47% (HR = 0.533). Additionally, subgroup analysis and Cox multivariate analysis showed that local treatment of liver metastases can also reduce the risk of death by 39%. For patient’s ineligible for local treatment, ICWM therapy was associated with significantly prolonged survival compared to those receiving standard care alone.

In recent years, Western medicine treatment for CRLM has continued to evolve, with multiple treatment strategies emerging: Chemotherapy combined with resection of liver metastases can prolong PFS (24.4 months vs 17.6 months, P = 0.03), but no OS benefit has been demonstrated[6]. Radiofrequency ablation and microwave ablation are safe and effective for small-volume metastases (≤ 3 cm), achieving local control rates of 92%-100% with low complication rates. However, their efficacy is limited for larger tumors, with 1-year and 2-year local control rates of 50%-95% and 45%-92%, respectively[7]. Compared with liver resection, ablation offers lower survival benefits, and even when combined with surgery, the survival advantage is only evident in the surgical group[8]. Transcatheter arterial chemoembolization combined with systemic therapy effectively treats chemotherapy-resistant unresectable CRLM, with a longer median PFS compared to patients receiving systemic chemotherapy alone (6.7 months vs 3.8 months; HR = 0.67, 95%CI: 0.49-0.91, P = 0.01), but there was no significant prolongation of mOS (18.4 months vs 14.8 months; HR = 0.92, 95%CI: 0.62-1.36; P = 0.67)[9]. In terms of medical treatment, targeted therapy combined with chemotherapy remains the mainstream approach, and no significant breakthroughs in survival benefits have been observed[10]. In summary, despite the continuous expansion of Western medicine treatment options for CRLM, there remains a significant bottleneck in improving treatment efficacy. Against this backdrop, ICWM might represent a potential alternative to enhance the current landscape of Western medicine treatment. However, current TCM research on CRLM primarily focuses on the in vivo mechanisms of action of individual herbal compounds[11] and classical formulas[12], with relatively limited high-quality clinical efficacy studies. Therefore, conducting in-depth clinical efficacy explorations of ICWM, such as the present study, holds significant value.

Based on the findings of this study, the ICWM demonstrates superior outcomes compared to standard Western medicine alone in improving survival benefits for patients with CRLM. CRC falls under the TCM categories of “intestinal fungus”, “accumulation”, and “intestinal disorder”. In the comprehensive management of advanced malignant tumors, our team proposes the TCM concept of “stage-based treatment”[13]. During different stages of treatment, such as chemotherapy and targeted therapy, we fully consider the impact of Western medicine treatment on the body’s balance between “Zheng” (positive energy) and “Xie” (negative energy), dynamically adjusting the ratio between “tonifying Zheng” and “eliminating Xie” to gradually restore the body to a state of “Yin Ping Yang Mi” (harmonious balance between Yin and Yang), thereby enabling patients to achieve benefits throughout the entire treatment process.

The evaluation indicators selected in this study are all objective and quantifiable, with minimal bias risk. However, as a retrospective study, it still has the following limitations: (1) This study only collected information on hospitalized patients. Patients who received only outpatient oral chemotherapy or targeted therapy were not included in the scope of this study; and (2) Due to severe deficiencies in safety information, this study was unable to evaluate the safety of ICWM treatment for CRLM.

CONCLUSION

In this cohort study, ICWM therapy was associated with extended OS and improved 3-year and 5-year survival rates in patients with CRLM. These findings provide preliminary real-world evidence supporting the potential clinical utility of an integrated approach in the management of CRLM. TCM and Western medicine improve the survival of malignant tumor patients through two distinct theoretical frameworks and diagnostic and therapeutic approaches, achieving complementary advantages. The observed survival outcomes suggest that such a combined strategy could represent a promising direction in comprehensive cancer care, warranting further investigation through rigorously designed prospective studies.