Chemotherapy plus bevacizumab vs chemoimmunotherapy for metastatic colorectal cancer: Real-world analysis

Abstract

BACKGROUND

Clinical trial evidence points to chemotherapy’s potential in augmenting the effects of immunotherapy.

AIM

To assess the effectiveness of first-line chemoimmunotherapy (CIT) for microsatellite stable (MSS) metastatic colorectal cancer (mCRC) verses standard-of-care (SOC; 5-fluorouracil/leucovorin/oxaliplatin/bevacizumab).

METHODS

This was a multicenter retrospective cohort study conducted in Peking University First Hospital and Jilin Cancer Hospital. Patients with MSS mCRC who had received first-line treatment were eligible. The Kaplan-Meier method and Cox proportional hazard model were used to evaluate progression-free survival (PFS) and to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). PFS was set as the primary endpoint. Propensity score (PS) was calculated to balance the baseline characteristics of the two cohorts. With PS, we performed three statistical methods, namely inverse probability weighting, PS matching, and additional adjustment for PS with multivariate cox regression.

RESULTS

Between July 2019 and November 2024, 148 eligible patients were enrolled, with 40 and 108 patients assigned to the CIT and SOC cohorts, respectively. At a global median follow-up of 21.4 months, the crude median PFS was 13.5 months (95%CI: 9.77-21.6) in the CIT cohort vs 9.1 months (95%CI: 7.8-10.6) in the SOC cohort, yielding a nonsignificant hazard ratio (HR) of 0.5645 (95%CI: 0.3637-0.8763, P = 0.01; SOC as reference). Multivariate Cox regression analysis, adjusted for sex, age > 60 years, Eastern Cooperative Oncology Group performance status, rat sarcoma mutation, primary tumor location (left vs right) and number of metastatic organs (liver/lung), demonstrated an adjusted HR of 0.55 (95%CI: 0.35-0.87, P = 0.011). PS-based analyses using PS matching (post-matching n = 40 vs 40), PS-adjusted multivariate Cox regression, and inverse probability weighting revealed consistent significant trends favoring CIT, with HRs for CIT of 0.5641 (95%CI: 0.3303-0.9635, P = 0.0361), 0.60 (95%CI: 0.38-0.96, P = 0.034), and 0.57 (95%CI: 0.337-0.973, P = 0.039), respectively.

CONCLUSION

Efficacy of CIT in MSS mCRC could surpass that of standard first-line chemotherapy. Further research is needed to investigate specific clinical characteristics or biomarkers to identify patients who may derive benefit from CIT.

Key Words: Microsatellite stable; Metastatic colorectal cancer; Immune checkpoint inhibitors; Programmed death 1

Core Tip: This study evaluated the effectiveness of chemoimmunotherapy (CIT) as the first-line regimen for microsatellite stable metastatic colorectal cancer vs standard-of-care (5-fluorouracil/leucovorin/oxaliplatin/bevacizumab) in the real world. The efficacy of CIT in microsatellite stable metastatic colorectal cancer could surpass that of standard first-line chemotherapy protocols. Further research is needed to investigate specific clinical characteristics or biomarkers to identify patients who may derive benefit from CIT.

INTRODUCTION

Colorectal cancer (CRC) ranks as the second deadliest malignancy worldwide[1] and carries a growing global socioeconomic burden[2]. Improving outcomes for patients with locally advanced rectal cancer and metastatic CRC (mCRC) remains a significant clinical challenge. mCRC patients face poor prognosis with 5-year survival rates < 15%[3]. Clinically, immune checkpoint inhibitors (ICIs) demonstrate remarkable efficacy in mCRC with high microsatellite instability (MSI-H) or deficient mismatch repair (dMMR). These tumors typically exhibit high tumor mutation burden, creating an inflammatory tumor microenvironment rich in tumor-infiltrating lymphocytes, particularly memory cells and cytotoxic T lymphocytes[4]. However, microsatellite stable (MSS) CRC patients derive minimal benefit from ICI monotherapy. The KEYNOTE-016 study reported objective response rates of 40% (4/10) and progression-free survival (PFS) of 78% (7/9) in dMMR CRC patients, compared to 0% (0/18) and 11% (2/18) respectively in MSS patients[5]. Similarly, KEYNOTE-028 found no clinical benefit in 22 of 23 programmed death-ligand 1 (PD-L1)-positive (PD-L1 ≥ 1%) CRC patients with MSS status[6]. Consequently, MSS/proficient mismatch repair (pMMR) CRC is considered an “immunologically cold” tumor, presenting major challenges for immunotherapy in advanced disease. Transforming these “cold” tumors into “hot” tumors through microenvironment modulation represents a crucial therapeutic goal[7]. Emerging evidence suggests some MSS/pMMR CRC patients may benefit from combination immunotherapy approaches. Various tumor-intrinsic microenvironmental factors contribute to ICI resistance in MSS/pMMR CRC, making immunosuppression modulation a primary target for combination strategies[7]. Current combination approaches include immunotherapy with chemotherapy, radiotherapy, targeted therapy, or dual immunotherapy. Chemoimmunotherapy (CIT) remains the most clinically utilized strategy, as chemotherapy can induce immunogenic cell death and counteract immune evasion, thereby potentiating immunotherapy effects[8]. Specifically, tumor cell destruction releases tumor antigens that activate host immune responses. ICI-chemotherapy combinations have become first-line standard for many solid tumors, with synergistic mechanisms including enhanced immunogenicity, reduced immunosuppression, increased immune cell infiltration, and immunological memory formation.

Several clinical trials are investigating CIT for MSS/pMMR CRC, although current results remain modest. The LEAP-017 trial evaluating lenvatinib plus pembrolizumab in pretreated non-MSI-H/dMMR mCRC failed to meet its overall survival (OS) primary endpoint[9]. The phase 2 CheckMate 9X8 study comparing nivolumab plus modified 5-fluorouracil/Leucovorin/oxaliplatin/bevacizumab with standard care showed potential benefit specifically in RAS mutant/MSS (CMS3 subtype) mCRC patients[10]. The BBCAPX study demonstrated enhanced response rates with sintilimab plus capecitabine and oxaliplatin (CapeOx)/bevacizumab in MSS mCRC patients with acceptable toxicity[11]. The METIMMOX phase 2 trial comparing fludarabine/oxaliplatin plus nivolumab with fludarabine/oxaliplatin alone in MSS mCRC reported median PFS of 6.6 vs 5.6 months[12]. While CIT offers promise for MSS/pMMR CRC patients, further validation of safety/efficacy profiles and optimization of personalized treatment strategies remain imperative.

Based on the aforementioned research, the synergistic combination of chemotherapy and anti-programmed death 1 (PD-1) immunotherapy should represent a critical research direction to transform current MSS mCRC treatment paradigms. To investigate the real-world performance of CIT as first-line treatment for MSS mCRC, we have commenced a multicenter retrospective cohort study assessing its safety and efficacy.

MATERIALS AND METHODS

Study design and participants

This study used a multicenter retrospective cohort research design. Patients with advanced CRC who were treated at Peking University First Hospital and Jilin Cancer Hospital between July 2019 and November 2024 were enrolled. The experimental group consisted of patients who received first-line treatment with chemotherapy combined with anti-PD-1 immunotherapy (CIT cohort), while the control group comprised patients who received conventional treatment (chemotherapy combined with bevacizumab) [standard-of-care therapy (SOC) cohort].

The inclusion criteria were: (1) Histologically or cytologically confirmed unresectable mCRC with measurable lesions based on the Response Evaluation Criteria in Solid Tumors; (2) pMMR or MSS, B-Raf proto-oncogene wild-type; (3) Patients receiving chemotherapy combined with anti-PD-1 immunotherapy, with or without bevacizumab, and those receiving chemotherapy combined with bevacizumab alone; (4) Having not undergone radiotherapy or completed radiotherapy > 4 weeks ago; and (5) Eastern Cooperative Oncology Group score ≤ 2.

The exclusion criteria were: (1) Patients with dMMR or MSI-H, or B-Raf proto-oncogene mutations; (2) Symptomatic brain metastases; (3) Uncontrolled active infection; (4) Dysphagia, intractable vomiting, or known drug absorption disorders; and (5) Patients with symptomatic or high-risk obstruction, bleeding, or perforation, or those who have undergone intestinal stent placement to relieve intestinal obstruction.

In most cases, the chemotherapy regimen comprised an oxaliplatin-based doublet (folinic acid, 5-fluorouracil, oxaliplatin, or CAPEOX) or a topoisomerase inhibitor (folinic acid, 5-fluorouracil, or irinotecan). Anti-PD-1 immunotherapy included penpulimab, pembrolizumab, sintilimab, tislelizumab, and toripalimab. The antiangiogenic agent was bevacizumab.

This study was conducted in compliance with the Declaration of Helsinki and approved by the Ethics Committee of Peking University First Hospital and Jilin Cancer Hospital. The requirement for patient approval or informed consent was waived by the Human Ethics Committee of Peking University First Hospital and Jilin Cancer Hospital, owing to the retrospective nature of the study and the use of anonymous clinical data for analysis.

Procedures

The study follow-up concluded in May 2025. The primary endpoint, PFS, was measured from enrollment to first disease progression or death. OS was calculated from enrollment to death, with censoring at last follow-up. The neutrophil-to-lymphocyte ratio (NLR) was determined by dividing the absolute neutrophil count by the absolute lymphocyte count (ALC). The lymphocyte-to-monocyte ratio was obtained by taking the quotient of ALC and the absolute monocyte count. Platelet-to-lymphocyte ratio (PLR) was calculated by dividing the absolute platelet count (PLT) by ALC. The systemic immune-inflammation index (SII), was calculated using the formula: PLT × absolute neutrophil count/ALC. Complete blood counts were collected at two time points: Baseline (within 14 days before initial anti-PD-1 therapy) and at 6 weeks post-treatment. The change in NLR (ΔNLR) was defined as the difference between the 6-week post-treatment value and the baseline value.

Statistical analysis

To minimize confounding bias, propensity score matching (PSM) was performed with a 1:1 ratio via the nearest-neighbor algorithm (MatchIt package[13]). Survival outcomes were estimated by Kaplan-Meier curves, while Cox regression models identified independent prognostic factors with corresponding hazard ratios (HRs) and 95% confidence intervals (CIs). All analyses were conducted using R version 4.4.2.

Sensitivity analysis

We conducted three sensitivity analyses to control for bias: Initial Schoenfeld residual testing (survival package[14]) confirmed proportional hazards assumption compliance, followed by appropriate univariate and multivariate regression modeling. Second, PSM analysis with varying matching ratios was applied to PFS and OS. Third, inverse probability of treatment weighting (IPTW) was used to adjust for baseline characteristics and evaluate treatment outcomes.

RESULTS

Patients

From July 2019 and November 2024, 148 eligible patients were enrolled. According to the treatment received, 40 and 108 patients were respectively allocated to the CIT cohort and SOC cohort. The study population included 91 males (61.5%) and 57 females (38.5%), with 72 patients (48.6%) aged > 60 years. Right-sided colon cancer accounted for 33.8% (n = 50), 81.6% (n = 120) had undergone primary tumor resection, and 39.9% (n = 59) presented with multiorgan metastases. All cases were pMMR, while RAS mutations were identified in 19% (n = 28) (Table 1).

Table 1 Demographic and clinical characteristics of patients at baseline.

Characteristics	Levels	n (%)
Age (year)	≤ 60	76 (51.4)
	> 60	72 (48.6)
Gender	Male	91 (61.5)
	Female	57 (38.5)
ECOG score	0-1	136 (91.9)
	2	12 (8.1)
Location	Right colon	50 (33.8)
	Left colon and rectum	98 (66.2)
Surgery	No	27 (18.4)
	Yes	120 (81.6)
No. of metastases	1	89 (60.1)
	≥ 2	59 (39.9)
Liver metastasis	No	64 (43.2)
	Yes	84 (56.8)
Lung metastasis	No	106 (71.6)
	Yes	42 (28.4)
RAS	Wild type	22 (14.9)
	KRAS mutation	26 (17.6)
	NRAS mutation	2 (1.4)
	Unknown	98 (66.2)
Group	SOC	108 (73.0)
	Chemoimmunotherapy	40 (27.0)

ECOG: Eastern Cooperative Oncology Group; SOC: Standard-of-care.

Effectiveness

With a median follow-up of 21.4 months (as of May 2025), the overall cohort demonstrated a median PFS of 10 months. The crude PFS was 13.5 (95%CI: 9.77-21.6) months in CI cohort vs 9.13 (7.8-10.6) months in SOC cohort. The PFS improvement achieved significance (HR = 0.56, 95%CI: 0.36-0.88, P = 0.011) (Figure 1A). However, no significant OS benefit was seen (HR = 0.61, 95%CI: 0.28-1.30, P = 0.197) (Figure 1B). Multivariate Cox regression analysis, adjusted for sex, age > 60 years, Eastern Cooperative Oncology Group performance status, RAS mutation, primary tumor location (left vs right) and number of metastatic organs (liver/Lung), demonstrated an adjusted HR of 0.55 (95%CI: 0.35-0.87, P = 0.011) in PFS (Table 2) and HR of 0.7 (95%CI: 0.32-1.52, P = 0.367) in OS (Table 3).

Figure 1 Kaplan-Meier curve of original cohort. A: Kaplan-Meier curve of original cohort (progression-free survival); B: Kaplan-Meier curve of original cohort (overall survival). SOC: Standard-of-care; CIT: Chemoimmunotherapy.

Table 2 Univariate and multivariate Cox analysis of the effect of prognostic factors in the original cohort (progression-free survival).

Dependent: Survival (PFS/30, status PFS)		All	HR (univariable)	HR (multivariable)
Gender	Male	90 (61.2)	-	-
	Female	57 (38.8)	0.83 (0.57-1.21, P = 0.335)	-
Age (year)	≤ 60	75 (51.0)	-	-
	> 60	72 (49.0)	1.02 (0.71-1.48, P = 0.915)	-
ECOG score	0-1	135 (91.8)	-	-
	2	12 (8.2)	1.68 (0.88-3.23, P = 0.117)	-
Primary tumor location	Right colon	49 (33.3)	-	-
	Left colon and rectum	98 (66.7)	1.11 (0.76-1.64, P = 0.587)	-
No. of metastatic organs	1	88 (59.9)	-	-
	≥ 2	59 (40.1)	1.39 (0.96-2.02, P = 0.082)	1.13 (0.76-1.67, P = 0.550)
Liver metastasis	No	64 (43.5)	-	-
	Yes	83 (56.5)	1.33 (0.92-1.94, P = 0.133)	-
Lung metastasis	No	105 (71.4)	-	-
	Yes	42 (28.6)	1.03 (0.68-1.56, P = 0.891)	-
RAS	Wild type	22 (15.0)	-	-
	Mutation	28 (19.0)	1.86 (0.98-3.53, P = 0.057)	1.95 (1.01-3.75, P = 0.045)
	Unknown	97 (66.0)	1.48 (0.87-2.53, P = 0.152)	1.65 (0.95-2.86, P = 0.076)
Group	Control group	107 (72.8)	-	-
	Experimental group	40 (27.2)	0.56 (0.36-0.88, P = 0.011)	0.55 (0.35-0.87, P = 0.011)

n = 148, events = 115, Likelihood ratio test = 12.91 on 4 df (P = 0.012). PFS: Progression-free survival; HR: Hazard ratio; ECOG: Eastern Cooperative Oncology Group.

Table 3 Univariate and multivariate Cox analysis of the effect of prognostic factors in the original cohort (overall survival).

Dependent: Survival (OS/30, status OS)		All, n (%)	HR (univariable)	HR (multivariable)
Gender	Male	90 (61.2)	-	-
	Female	57 (38.8)	0.74 (0.41-1.34, P = 0.316)	0.94 (0.51-1.74, P = 0.848)
Age (year)	≤ 60	75 (51.0)	-	-
	> 60	72 (49.0)	1.25 (0.71-2.19, P = 0.434)	0.98 (0.55-1.74, P = 0.942)
ECOG score	0-1	135 (91.8)	-	-
	2	12 (8.2)	1.03 (0.37-2.87, P = 0.957)	-
Primary tumor location	Right colon	49 (33.3)	-	-
	Left colon and rectum	98 (66.7)	1.00 (0.56-1.78, P = 0.992)	-
No. of metastatic organs	1	88 (59.9)	-	-
	≥ 2	59 (40.1)	1.49 (0.85-2.60, P = 0.164)	-
Liver metastasis	No	64 (43.5)	-	-
	Yes	83 (56.5)	3.55 (1.88-6.69, P < 0.001)	3.48 (1.80-6.72, P < 0.001)
Lung metastasis	No	105 (71.4)	-	-
	Yes	42 (28.6)	0.73 (0.38-1.39, P = 0.335)	-
RAS	Wild type	22 (15.0)	-	-
	Mutation	28 (19.0)	0.92 (0.41-2.07, P = 0.842)	-
	Unknown	97 (66.0)	0.63 (0.31-1.27, P = 0.195)	-
Group	Control group	107 (72.8)	-	-
	Experimental group	40 (27.2)	0.61 (0.28-1.30, P = 0.197)	0.70 (0.32-1.52, P = 0.367)

n = 148, events = 51, Likelihood ratio test = 17.74 on 1 df (P < 0.001). OS: Overall survival; HR: Hazard ratio; ECOG: Eastern Cooperative Oncology Group.

Sensitivity analysis

With PS we performed three statistical methods, namely PSM (sample size was 40 vs 40 after matching); additional adjustment for PS with multivariate Cox regression; and inverse probability weighting. Subgroup analysis revealed a longer median PFS in the CIT group (13.5 months) vs the SOC group (9.1 months), although this difference was not significant (HR = 0.564, 95%CI: 0.330-2.964, P = 0.036) (Figure 2A). Adjusting the PSM ratio did not alter the final PFS (Figure 2B-D). However, the difference in OS did not reach significance (HR = 1.067, 95%CI: 0.418-2.729, P = 0.89) (Figure 3A). Adjusting the PSM ratio did not affect the final OS (Figure 3B-D). After additional adjustment for PS with multivariate Cox regression, the HR of CIT was 0.60 (95%CI: 0.38-0.96, P = 0.034) for PFS (Table 4) and the HR of CIT was 0.74 (95%CI: 0.33-1.65, P = 0. 468) for OS (Table 5). IPTW analysis corroborated this finding for PFS (P = 0.021) (Figure 4A) and OS (P = 0.703) (Figure 4B).

Figure 2 Progression-free survival sensitivity analysis. A: After propensity score matching (PSM) analysis of progression-free survival (PFS) (ratio = 1); B: After PSM analysis of PFS (ratio = 2); C: After PSM analysis in PFS (ratio = 3); D: After PSM analysis of PFS (ratio = 4). SOC: Standard-of-care; CIT: Chemoimmunotherapy.

Figure 3 Overall survival sensitivity analysis. A: After propensity score matching (PSM) analysis of overall survival (OS) (ratio = 1); B: After PSM analysis of OS (ratio = 2); C: After PSM analysis of OS (ratio = 3); D: After PSM analysis of OS (ratio = 4). SOC: Standard-of-care; CIT: Chemoimmunotherapy.

Figure 4 Inverse probability of treatment weighting analysis. A: After inverse probability of treatment weighting analysis (progression-free survival); B: After inverse probability of treatment weighting analysis (overall survival). SOC: Standard-of-care; CIT: Chemoimmunotherapy.

Table 4 Propensity score-adjusted multivariable cox regression analysis of prognostic factors associated with progression-free survival in the original cohort.

Dependent: Survival (PFS, status PFS)		All, n (%)	HR (univariable)
Gender	Male	90 (61.2)	-
	Female	57 (38.8)	1.18 (0.20-7.13, P = 0.855)
Age (year)	≤ 60	75 (51.0)	-
	> 60	72 (49.0)	1.07 (0.68-1.69, P = 0.757)
ECOG score	0-1	135 (91.8)	-
	2	12 (8.2)	1.96 (0.95-4.04, P = 0.070)
Primary tumor location	Right colon	49 (33.3)	-
	Left colon and rectum	98 (66.7)	1.34 (0.72-2.50, P = 0.356)
No. of metastatic organs	1	88 (59.9)	-
	≥ 2	59 (40.1)	0.73 (0.08-6.57, P = 0.782)
Liver metastasis	No	64 (43.5)	-
	Yes	83 (56.5)	1.27 (0.83-1.92, P = 0.268)
Lung metastasis	No	105 (71.4)	-
	Yes	42 (28.6)	0.98 (0.61-1.60, P = 0.946)
RAS	Wild type	22 (15.0)	-
	Mutation	28 (19.0)	2.17 (1.07-4.38, P = 0.031)
	Unknown	97 (66.0)	1.76 (0.99-3.15, P = 0.056)
Group	Control group	107 (72.8)	-
	Experimental group	40 (27.2)	0.60 (0.38-0.96, P = 0.034)
PS	mean ± SD	0.3 ± 0.1	0.11 (0.00-9673.56, P = 0.702)

n = 147, events = 114, Likelihood ratio test = 18.9 on 11 df (P = 0.063). PFS: Progression-free survival; HR: Hazard ratio; ECOG: Eastern Cooperative Oncology Group; PS: Propensity score.

Table 5 Propensity score-adjusted multivariable cox regression analysis of prognostic factors associated with overall survival in the original cohort.

Dependent: Survival (OS, status OS)		All, n (%)	HR (multivariable)
Gender	Male	90 (61.2)	-
	Female	57 (38.8)	0.32 (0.02-5.24, P = 0.421)
Age (year)	≤ 60	75 (51.0)	-
	> 60	72 (49.0)	1.10 (0.55-2.23, P = 0.782)
ECOG score	0-1	135 (91.8)	-
	2	12 (8.2)	0.92 (0.31-2.77, P = 0.880)
Primary tumor location	Right colon	49 (33.3)	-
	Left colon and rectum	98 (66.7)	1.47 (0.59-3.64, P = 0.404)
No. of metastatic organs	1	88 (59.9)	-
	≥ 2	59 (40.1)	3.78 (0.14-99.03, P = 0.424)
Liver metastasis	No	64 (43.5)	-
	Yes	83 (56.5)	3.16 (1.53-6.53, P = 0.002)
Lung metastasis	No	105 (71.4)	-
	Yes	42 (28.6)	0.81 (0.35-1.84, P = 0.610)
RAS	Wild type	22 (15.0)	-
	Mutation	28 (19.0)	0.91 (0.36-2.26, P = 0.833)
	Unknown	97 (66.0)	0.58 (0.27-1.26, P = 0.168)
Group	Control group	107 (72.8)	-
	Experimental group	40 (27.2)	0.74 (0.33-1.65, P = 0.468)
PS	mean ± SD	0.3 ± 0.1	719.14 (0.00-25553659310.55, P = 0.458)

n = 147, events = 50, Likelihood ratio test = 21.98 on 11 df (P = 0.025). OS: Overall survival; HR: Hazard ratio; ECOG: Eastern Cooperative Oncology Group; PS: Propensity score.

Exploratory biomarker analysis

We stratified patients into two groups based on post-treatment NLR changes: Decreased (ΔNLR < 0) and increased (ΔNLR ≥ 0), to evaluate the potential of ΔNLR as a biomarker for MSS mCRC patients receiving PD-1 inhibitor therapy. Additionally, we investigated the predictive value of dynamic changes in SII, lymphocyte-to-monocyte ratio (LMR), and PLR for treatment efficacy. The results demonstrated that increased post-treatment NLR (ΔNLR ≥ 0), elevated SII (≥ 0), and lower LMR (≥ 0) were associated with poorer clinical outcomes (Figure 5).

Figure 5 Kaplan-Meier curves for progression-free survival. Progression-free survival according to alteration of neutrophil-to-lymphocyte ratio. A: After two cycles of immunotherapy; B: Systemic immune-inflammation index alteration; C: Lymphocyte-to-monocyte ratio alteration; D: Platelet-to-lymphocyte ratio alteration. HR: Hazard ratio; CI: Confidence interval; NLR: Neutrophil-to-lymphocyte ratio; SII: Systemic immune-inflammation index; LMR: Lymphocyte-to-monocyte ratio; PLR: Platelet-to-lymphocyte ratio.

DISCUSSION

CRC represents a significant global health challenge, with persistently high incidence and mortality rates. In China, this poses an equally severe public health burden. mCRC demonstrates particularly disappointing 5-year survival rates. The MSS subtype accounts for the majority of mCRC cases, yet treatment efficacy and options remain limited for these patients, representing a substantial unmet clinical need. While traditional chemotherapy serves as the cornerstone, its potential is constrained by toxicity and drug resistance issues. Immunotherapy has achieved remarkable efficacy in MSI-H/dMMR advanced CRC, benefiting 5%-10% of patients. However, for > 90% of MSS/pMMR patients, current immunotherapy monotherapy shows limited efficacy, characterizing these as immunologically cold tumors. Our long-term research focus has been on transforming these cold tumors into hot tumors to expand immunotherapy benefits for MSS patients[15].

The IBI363 study enrolled 68 patients (monotherapy arm) and 73 patients (combination arm) with mCRC. Baseline characteristics were balanced: MSS/pMMR (87%/90%), liver metastases (62%/55%), RAS mutations (43%/41%), ≥ 3 prior lines of therapy (63%/53%), and prior immunotherapy exposure (28%/16%). Treatment regimens included IBI363 monotherapy and combination with bevacizumab. With median follow-up of 11.8 (0.4-22.5) and 5.1 (1.2-14.9) months, respectively, efficacy analysis demonstrated that IBI363 monotherapy significantly improved OS compared to historical standard treatment data. The combination regimen showed particularly encouraging efficacy with acceptable safety, warranting further investigation[16]. A French multicenter phase 2 randomized trial compared atezolizumab plus modified docetaxel-cisplatin-5-fluorouracil vs modified docetaxel-cisplatin-5-fluorouracil alone in metastatic/advanced anal squamous cell carcinoma. Both arms showed favorable efficacy and safety, with 1-year PFS of 44.2% in the combination arm, though the primary endpoint was not met[17,18]. The current treatment focus of MSS mCRC mainly includes immune combination therapies, new drug development, and biomarker exploration[7,19]. MSS CRC exhibits a poor response to immunotherapy. Several strategies for exploring the application of immunotherapy in this context are worthy of attention. Firstly, different combination strategies. Immuno-combination therapies are currently a hotspot of exploration and a promising direction. For instance, the strategy of combining IBI363 with the antiangiogenic drug bevacizumab, or immunotherapy combined with chemotherapy, other immunomodulators (such as cytotoxic T lymphocyte-associated protein-4 inhibitors), or targeted agents, can synergistically enhance the therapeutic effect by altering the tumor microenvironment[16]. For example, the BBCAPX study[11], which investigated sintilimab in combination with chemotherapy and bevacizumab, demonstrated encouraging early data in the MSS population, and the ongoing phase 3 study is highly anticipated. Secondly, the development of novel immunotherapeutic drugs. Examples include bispecific antibodies, tumor vaccines, cell therapies such as chimeric antigen receptor T-cell immunotherapy/T-cell-receptor-engineered T-cell therapy, and oncolytic viruses. These novel agents, through their unique mechanisms of action, can directly kill tumor cells or more effectively activate the immune system, potentially offering new options for the treatment of MSS mCRC. Thirdly, the exploration of more precise biomarkers. The future lies in identifying MSS patient subgroups most likely to benefit from specific immunotherapies or combination strategies. This requires a deeper understanding of the characteristics of the tumor microenvironment and the discovery of reliable predictive biomarkers to better achieve individualized precision medicine.

NLR, LMR, and PLR are closely related to systemic inflammation. These inflammatory markers are also used as prognostic biomarkers in various malignancies, including esophageal cancer[20], lung cancer[21], and melanoma[22], undergoing immunotherapy. A limited number of previous studies have explored the dynamic changes in NLR at different time points during immunotherapy for lung cancer. Nakaya et al[23] reported that in non-small cell lung cancer (NSCLC) patients, an increase in NLR after one cycle of nivolumab treatment was significantly associated with shorter PFS and OS, whereas baseline NLR was not associated with prognosis in NSCLC patients. Another small-sample study (n = 19) showed that NSCLC patients with an NLR increase of > 30% after one cycle of immunotherapy had a significantly shorter time to treatment failure[24]. These findings suggest that early changes in NLR during immunotherapy may be a potential predictive marker for disease progression.

The SII, an immunological and inflammatory marker in peripheral blood, to some extent reflects the balance between the body’s immune and inflammatory responses. In the tumor microenvironment, inflammatory markers and cells participate in tumor initiation and progression through interactions with the cellular matrix. When the body experiences an inflammatory response or a decline in immunity, an increase in neutrophils or a decrease in lymphocytes indicates an imbalance between immunity and inflammation, leading to abnormalities in the tumor immune microenvironment, which may result in immunotolerance or immune escape, and facilitate tumor progression or metastasis. Platelets contribute to the creation of a microenvironment conducive to tumor metastasis by controlling the release of various growth factors[25,26]. As a comprehensive inflammatory index integrating neutrophil, lymphocyte, and PLTs, SII can reflect the systemic immune-inflammatory status of patients. Recent studies have suggested that an elevated SII is associated with poor survival outcomes in CRC[27]. The results of this study indicate that SII is a factor influencing PFS in patients. An elevated SII is associated with shorter PFS in MMR-proficient advanced CRC patients receiving immunotherapy combined with chemotherapy.

We recognize that our study had some limitations. Firstly, the findings necessitate validation through larger studies. The maturity of the PFS and OS data in our retrospective assessment was low. Additionally, patient management practices were not as standardized, which may have introduced certain biases into the results. Secondly, our study did not evaluate PD-L1 expression in the patients. Consequently, we were unable to analyze the correlation between PD-L1 expression levels and the efficacy of anti-PD-1 immunotherapy. Thirdly, there was significant heterogeneity in the treatment regimens. This included the use of multiple chemotherapy protocols, such as folinic acid, 5-fluorouracil, oxaliplatin, or CAPEOX, folinic acid, 5-fluorouracil, irinotecan, and CAPEOX, as well as five different anti-PD-1 agents. Such heterogeneity could potentially confound the comparisons of treatment efficacy.

CONCLUSION

CIT shows potential advantages over conventional first-line chemotherapy for MSS mCRC, although additional studies are required to determine predictive biomarkers and identify patient subgroups most likely to benefit.

ACKNOWLEDGEMENTS

We are grateful to the patients and their families for supporting the study.