Meta-analysis of the efficacy of neoadjuvant immunotherapy combined with radiotherapy and chemotherapy for locally advanced rectal cancer

Abstract

BACKGROUND

Immunotherapy is an approved treatment for metastatic rectal cancer in patients with defective mismatch repair (MMR).

AIM

To examine the clinical efficacy of neoadjuvant immunotherapy combined with radiotherapy and chemotherapy for the treatment of locally advanced rectal cancer (LARC), with a focus on patients with proficient MMR (pMMR) and microsatellite stability.

METHODS

Two researchers searched multiple databases for publications up to September 2024. All included publications examined neoadjuvant immunotherapy for LARC, and reported major pathological response (MPR), pathological complete response (pCR), clinical complete response (CCR), and rates of R0 resection and anus-preserving surgery. Meta-analysis, subgroup analysis, sensitivity analysis, and analysis of publication bias were performed.

RESULTS

We included 15 publications (796 patients). The MPR, pCR, and CCR were significantly better in the group that received immunotherapy (all P < 0.05), especially for patients with pMMR. In addition, the rate of R0 resection and anus-preserving surgery were also significantly greater in the group that received neoadjuvant immunotherapy (both P < 0.05). Hematological toxicity and abnormal liver function were the most common clinical adverse events above grade 3. Most patients successfully completed the immunotherapy treatment. The incidence of immune-related adverse reactions was 0%-13.5%, and the severities of these events were generally considered acceptable.

CONCLUSION

The addition of neoadjuvant immunotherapy improved the clinical remission rate of patients who had LARC with pMMR, and the treatment-related adverse reactions were generally acceptable. Neoadjuvant immunotherapy combined with radiotherapy and chemotherapy should be considered for patients with LARC.

Key Words: Immunotherapy; Defective mismatch repair; Meta-analysis; Locally advanced rectal cancer; Proficient mismatch repair; Microsatellite stability

Core Tip: Immunotherapy is an approved treatment for metastatic rectal cancer in patients with defective mismatch repair (MMR). Neoadjuvant chemotherapy combined with immunotherapy can benefit patients who have locally advanced rectal cancer (LARC) with proficient MMR (pMMR)/microsatellite stability (MSS). Through a meta-analysis with 15 publications included, the results of our meta-analysis showed that neoadjuvant immunotherapy combined with radiotherapy and chemotherapy can improve the clinical remission rate (major pathologic response, clinical complete response, and pathologic complete response) of patients with LARC, especially for those with pMMR/MSS, and this treatment did not increase the incidence of postoperative complications and adverse reactions.

INTRODUCTION

The International Agency for Research on Cancer of the World Health Organization reported in 2022 that there were about 1.926 million newly diagnosed cases and 904000 deaths from colorectal cancer (CRC) worldwide. Thus, CRC ranked among the top five malignancies in terms of morbidity and mortality[1]. In 2020, China had an estimated 517000 newly diagnosed cases and 286000 deaths from CRC, making it a serious threat to human health[2]. The concept of locally advanced rectal cancer (LARC) is relatively broad and includes a heterogeneous group of patients. At present, the standard treatment for LARC is neoadjuvant chemoradiotherapy (NACRT) followed by total mesorectal resection (TME). Application of this treatment for LARC can provide an objective response rate (ORR) of 40% and a pathologic complete response (pCR) rate of 14% to 28%[3,4].

Although regimens using chemoradiotherapy and surgery have been used to treat LARC for nearly 20 years, two practical problems remain. First, after this treatment patients have a high rate of distant metastasis (up to 30% in five years) and a poor survival rate. Second patients with low rectal cancer typically require permanent colostomy, and this greatly reduces quality of life. Therefore, it is important to reduce the rate of distant metastasis and increase the rate of complete remission in patients with LARC.

The emergence and widespread adoption of different immunotherapy regimens for multiple types of cancer has led to its use in certain patients with LARC to decrease the risk of metastasis. In contrast to systemic chemotherapy, immune checkpoint inhibitors (ICIs) do not directly kill tumor cells; instead, they block the inhibitory signaling pathway between T lymphocytes and antigen presenting cells, activate tumor-specific T cells so they migrate to the tumor, re-establish the tumor immune microenvironment, and restore the normal anti-tumor immune response of the body. The KEYNOTE-016 study showed that patients who had metastatic CRC with defective mismatch repair (MMR)/microsatellite instability-high (MSI-H) received significant benefit from monoclonal antibody immunotherapy that targeted the programmed cell death-ligand 1 (PD-L1)[5], and clinical guidelines now recommend immunotherapy for these patients. However, only about 5% of patients with CRC have defective MMR (dMMR)/MSI-H, and the first choice for other patients, i.e., those with proficient MMR (pMMR) and microsatellite stability (MSS)] is still radiotherapy and chemotherapy.

In 2020, Yuki et al[6] described a new strategy of adjuvant therapy consisting of sequential immunotherapy after long-term radiotherapy and chemotherapy for CRC. They reported a pCR of 60% in patients with MSI-H status and 30% in patients with MSS status. This suggests that many CRC patients with pMMR/MSS and all patients with dMMR/MSI-H may benefit from treatment with nivolumab [which targets the programmed cell death 1 (PD-1) receptor] plus ipilimumab (which targets a protein that down-regulates immune responses)[7]. Subsequently, a number of prospective studies showed that immunotherapy with neoadjuvant radiotherapy and chemotherapy increased the rate of complete remission in patients with LARC[8,9]. However, there are still some unresolved problems with these “immune +” treatment modalities.

Because there is evidence that neoadjuvant chemotherapy combined with immunotherapy can benefit patients who have LARC with pMMR/MSS, we performed a literature review and meta-analysis of publications that examined the effect of this regimen to assess its safety and efficacy as a treatment for LARC.

MATERIALS AND METHODS

Retrieval of publications

The databases of PubMed, Embase, Cochrane Library, China HowNet, and Wanfang were searched to identify eligible publications as of September 2024. The search terms were: [Rectal cancer or rectal tumor or immunotherapy] or [immunotherapy or ICI or immune checkpoint blocker and neoadjuvant therapy] and [neoadjuvant chemotherapy or perioperative treatment or perioperative radiotherapy or chemotherapy]. All eligible publications were in the English or Chinese language. The PROSPERO website registration number was CRD42025638933.

Inclusion and exclusion criteria

All included studies: (1) Examined patients with LARC; (2) Were RCTs or prospective cohort studies; (3) Included patients undergoing neoadjuvant immunotherapy; (4) Provided complete and original survival data for quantitative calculations; (5) Had complete text available; and (6) Were written in the Chinese or English language. Studies were excluded if they: (1) Examined patients with metastatic CRC; (2) Were case reports, meeting reports, or letters; (3) Did not administer adjuvant immunotherapy; (4) Had incomplete data or insufficient clinical details; or (5) Were duplicate publications.

Different studies had different criteria about the definition of LARC. The commonly used MERCURY research defines LARC as: T3c, T3d or T4 rectal cancer. EXPERT study is defined as rectal cancer with T1-4 N2, low T3 and T4 and suspicious positive circumferential margin. The German standard is: Any rectal cancer with T, N+ or T3 or T4.

Data extraction and quality assessment

Two researchers (Yan WX and Xiong ZY) reviewed the abstract and full text of each publication identified in the search, and used the inclusion and exclusion criteria to identify eligible studies. If there was a disagreement about the eligibility of a publication, a third researcher (Li N) was consulted to reach a resolution. If the inclusion criteria were met, the two researchers read the full text to determine whether it was eligible. The extracted contents included: Basic information, including baseline data (country, publication year, number of patients, age, sex, TNM stage, tumor location, type and sequence of radiotherapy and chemotherapy, PD-L1 score, MMR status), main results [major pathological response (MPR), clinical complete response (CCR), and pCR] and secondary results (adverse effects, rate of R0 resection, and rate of anus-preserving surgery). Because not all the included studies were randomized controlled studies, the ROBINS-I tool was used to evaluate study quality. The therapies described in each study were independently evaluated by two researchers (Yan WX and Yuan HQ). If they had a difference of opinion regarding the eligibility of a study, a third researcher (Li N) was consulted to reach a resolution. All included studies had high quality and were suitable for a systematic meta-analysis.

Statistical analysis

The primary endpoints were MPR, CCR, and pCR, and the secondary endpoints were adverse effects, rate of R0 resection, and rate of anus-preserving surgery. Statistical analysis was performed using Stata version 11.0 and RevMan version 5.3. The odds ratio (OR) and relative risk (RR) were reported with 95% confidence intervals, and a difference was considered significant if the P value was less than 0.05. The Q test and I² test were used to evaluate the heterogeneity of included studies, and the results were used to determine which statistical test should be used for comparisons. If P was greater than 0.1 or I² was less than 50%, the studies were considered homogeneous and a fixed-effect model was used for analysis. If P value was less than 0.1 or I² was 50% or more, the studies were considered heterogeneous, and a random effect model was used for analysis. If the heterogeneity of studies was statistically significant, the possible reasons for this heterogeneity were examined by subgroup analysis. Funnel plots were also used to assess publication bias, and a P value below 0.05 in Egger’s test was considered significant.

RESULTS

Literature retrieval

We initially retrieved 494 publications, and an initial analysis indicated that 406 of these publications were unsuitable. After reading the title, abstract, full text, and application of the inclusion and exclusion criteria, we ultimately included 15 publications (Figure 1). Application of the ROBINS-I tool indicated that the quality of all 15 studies had high quality and was suitable for systematic meta-analysis (Figure 2).

Figure 1  Study selection based on PRISMA criteria.

Figure 2 Risk of bias (ROBINS-I) in assessment of major pathological response, pathological complete response, and clinical complete response. A: Major pathological response; B: Pathological complete response; C: Clinical complete response.

Basic characteristics of studies

One publication was in the Chinese language, and the other 14 were in the English language. All 15 studies were published from 2020 to 2024. These studies examined 796 patients with LARC who received neoadjuvant immunotherapy combined with radiotherapy and chemotherapy. The studies were performed in China, Europe, North America, or elsewhere, and each study examined 11 to 121 patients (Table 1).

Table 1 Basic information of the 796 patients with locally advanced rectal cancer in the 15 included publications.

Ref.	Country	Stage	Patient (n)	Age (years)	Sex (male/female)	TNM status	Median distance from anal verge (cm)	PD-L1 (status/score)	MMR/MSS (n)
Gao et al[14], 2022	China	II	11	57.1	9/2	cT3-4N0M0 or cT1-4N+M0	5.1 ± 2.1	NR	pMMR/MSS
Rahma et al[15], 2021	United States	II	90	55.7	60/30	T3-4/N+	NR	NR	NR
Zhou et al[41], 2022	China	II	23	55	14/9	II/III	1	NR	pMMR/MSS
Grassi et al[17], 2023	Italy	II	55	63	27/28	T3-4/N+	5	NR	dMMR: 2 L pMMR: 47; unknown: 6
Shamseddine et al[18], 2020	Lebanon	II	36	62.2	9/4	II/III	10	NR	pMMR/MSS
Wu et al[19], 2022	China	II	25	58	19/6	T3-4/N+	5.3	NR	pMMR/MSS
Xiao et al[9], 2024	China	II	67	-	-	II/III	< 5: 37.3%; ≥ 5, < 10: 56.7%; ≥ 10: 6%	NR	pMMR/MSS
Lin et al[20], 2021	China	II	30	57	17/13	T3-4/N+	4.7	(-): 66.7%; ≥ 1: 20%; unknown: 13.3%	dMMR: 1; pMMR: 28; unknown: 1
Lin et al[21], 2024	China	III	113	58	75/38	II/III	≤ 5: 49.6%; > 5: 50.4%	< 1: 14.2%; ≥ 1: 55.8%	pMMR: 42/105
Bando et al[22], 2022	Japan	IB	44	60.6	29/15	II/III	7	TPS ≥ 1: 75% (6/8); TPS < 1: 17% (5/30)	MSS: 39; MSI-H: 5
George et al[23], 2022	United States	II	45	NR	NR	II/III	NR	NR	NR
Salvatore et al[24], 2021	Italy	II	101	63	62/39	II/III	NR	NR	NR
Carrasco et al[25], 2023	North America	IB/II	26	66	13/13	II/III	NR	NR	NR
Xia et al[42], 2024	China	II	121	57	NR	II/III	< 5: 80.2%	NR	pMMR: 122; A group: 62; B group: 59
Wang et al[26], 2025	China	II	32	NR	23/9	T3/4	< 5: 40.6%	NR	pMMR/MSS

PD-L1: Programmed cell death-ligand 1; MMR: Mismatch repair; MSS: Microsatellite stability; pMMR: Proficient mismatch repair; NR: Not reported; dMMR: Defective mismatch repair; TPS: Tumor proportion score.

Analysis of MPR, CCR, and pCR

We analyzed the MPR, pCR, and CCR of patients enrolled in different treatment groups using a random effects model because the studies had significant heterogeneity (Table 2). The results show that the addition of immunotherapy to patients with LARC who received neoadjuvant radiotherapy and chemotherapy led to improvements of MPR (Figure 3A; I² = 80%, τ² = 0.0242, P < 0.01), pCR (Figure 3B; I² = 68%, τ² = 0.0084, P < 0.01), CCR (Figure 3C; I² = 91%, τ² = 0.0185, P < 0.01), and overall complete response, defined as pCR combined with CCR (Figure 3D; I² = 70%, τ² = 0.0104, P < 0.01).

Figure 3 Forest plot of rate risk ratio. A: Major pathological response; B: Pathological complete response (pCR); C: Clinical complete response (CCR); D: PCR + CCR.

Table 2 Outcomes of the 796 patients with locally advanced rectal cancer in the 15 included publications (%).

Ref.	MPR	pCR	CCR	CR	R0 resection	Sphincter preservation	Surgical complications	Adverse effects, grade 3 or higher
Gao et al[14], 2022	27.30	54.50	18.20	54.50	100.0	100.0	NR	9.10
Rahma et al[15], 2021	NR	31.9	13.9	32	94	59.4	NR	48.2
Zhou et al[41], 2022	NR	20	43.5	25.2	NR	95.5	NR	17.4
Grassi et al[17], 2023	36.30	34.50	10.9	32.7	96.4	NR	NR	18
Shamseddine et al[18], 2020	67.5	37.5	NR	37.5	92.3	NR	NR	25
Wu et al[19], 2022	33.3	33.30	16	44	100	66.70	14.30	36
Xiao et al[9], 2024	NR	31.30	13.40	44.80	NR	94.70	NR	49.4
Lin et al[20], 2021	66.60	48.1	11.1	59.2	100	88.9	14.80	26.7
Lin et al[21], 2024	NR	39.8	1.7	41.6	96.2	94.2	10.5	16.8
Bando et al[22], 2022	38.80	31.85	2.70	30	pMMR: 70.2; dMMR: 80	pMMR: 87; dMMR: 100	31.50	18.2
George et al[23], 2022	NR	22.20	NR	31.1	NR	NR	NR	NR
Salvatore et al[24], 2021	60	23	NR	23	NR	NR	NR	8
Carrasco et al[25], 2023	NR	27	NR	24	NR	NR	NR	49
Xia et al[42], 2024	A group: 67.5; B group: 70.6	50	Total: 39.7; A group: 43.5; B group: 35.6	Total: 55.4; A group: 56.5; B group: 54.2	NR	A group: 82.3; B group: 86.4	NR	A group: 24.2; B group: 33.9
Wang et al[26], 2025	75	62.50	NR	NR	NR	87.50	NR	NR

MPR: Major pathological response; pCR: Pathological complete response; CCR: Clinical complete response; CR: Complete response; pMMR: Proficient mismatch repair; NR: Not reported; dMMR: Defective mismatch repair.

We then performed subgroup analysis of MPR, pCR, and CCR in patients with pMMR status. The results showed that the group which received immunotherapy had an improved MPR (Figure 4A; I² = 72%, τ² = 0.0207, P < 0.01), pCR (Figure 4B; I² = 72%, τ² = 0.0108, P < 0.01), CCR (Figure 4C; I² = 91%, τ² = 0.0193, P < 0.01). In addition, patients who received immunotherapy were more likely to receive R0 resection (Figure 4D; I² = 58%, τ² = 0, P = 0.01) and anus-preserving surgery (Figure 4E; I² = 83%, τ² = 0.0119, P < 0.01).

Figure 4 Subgroup analysis of major pathological response, pathological complete response, and clinical complete response in patients with proficient mismatch repair status. A-C: Forest plot of major pathological response rate risk ratio in the subgroup of patients with proficient mismatch repair (pMMR; A), pathological complete response rate risk ratio in the subgroup of patients with pMMR (B), and clinical complete response rate risk ratio in the subgroup of patients with pMMR (C); D and E: Forest plot of R0 resection (D) and the sphincter preservation (E) rate risk ratio.

Safety analysis

All 15 studies reported the incidence of adverse reactions ≥ 3 (data not shown), and the OR was used for analysis. OR > 1 indicated that the incidence of adverse events was high, and the difference was statistically significant (P < 0.05). Hematological toxicity (incidence rate: 8%) and abnormal liver function (incidence rate: 50%) were the most common grade 3 and above adverse events. The incidence rate of immune-related adverse reactions ranged from 0% to 13.5%, and endocrine functional damage was common, but the severity was generally considered acceptable.

Sensitivity analysis

Sensitivity analysis showed that the exclusion of any single study had no obvious impact on the overall results (data not shown), indicating that the meta-analysis results are stable and reliable. After excluding each study, we need to recalculate the I² statistic, observe the trend of heterogeneity indicators, and judge whether there are studies that have made great contributions to heterogeneity.

Publication bias

We used funnel plots to assess the presence of publication bias (Figure 5), and used Egger’s test (threshold: P < 0.05) to determine the significance of bias. The results indicated there was no evidence of significant publication bias.

Figure 5 Funnel plot analysis of publication bias for major pathological response, pathological complete response, and clinical complete response. A: Major pathological response; B: Pathological complete response; C: Clinical complete response.

DISCUSSION

Neoadjuvant radiotherapy and chemotherapy of patients with LARC can provide a local recurrence rate as low as 5% to 7%, an overall survival rate of 65% to 75%, and a decreased rate of local recurrence, but the effect of this treatment on long-term prognosis is uncertain[10-13]. The important clinical goals for the treatment of these patients are improving the anus-preserving rate, increasing the complete remission rate and then improving the long-term survival. Recent prospective clinical studies of LARC showed that neoadjuvant chemotherapy combined with immunotherapy increased the rate of clinical remission and anus preservation in patients with low rectal cancer, and also significantly increased the proportion of patients who elected a watchful waiting approach[14-26]. Our meta-analysis identified publications that examined neoadjuvant chemotherapy and radiotherapy combined with immunotherapy for LARC, and analyzed the clinical and therapeutic benefit of neoadjuvant immunotherapy, with a focus on patients with pMMR/MSS. The current results confirmed our initial hypothesis. In particular, adding immunotherapy to neoadjuvant radiotherapy and chemotherapy for treatment of LARC significantly improved the pCR (RR = 0.348) and CCR (RR = 0.192), increased the rate of anus preservation (RR = 0.863), and the overall toxicity (immune-related adverse reactions ranged from 0% to 13.5%) that was within an acceptable range.

PD-1 is highly expressed on activated T lymphocytes, and its binding to PD-L1 (a transmembrane protein in tumor cells) inhibits the immune function of effector T cells[27]. Radiotherapy can also increase the activity of cytotoxic T cells and the generation and presentation of antigens, thus increasing the anti-tumor immune response of ICIs[28] due to synergistic effects[29]. About 45% of the tumor microenvironment of MSS-type CRC has high infiltration of CD3+ CD8+ T cells[30], and some of these patients may respond to immunotherapy[31,32]. Thus, the PANDORA trial examined the efficacy of concurrent chemoradiotherapy combined with neoadjuvant immunotherapy for LARC, and reported a pCR of 34.5% with tolerable toxicity[17]. The PKUCH 04 trial evaluated the efficacy of total neoadjuvant therapy (TNT) combined with neoadjuvant immunotherapy for pMMR/MSS rectal cancer and reported the pCR of 21 patients undergoing surgery was 33.3%[19]. The NRG-GI002 trial compared the efficacy and safety of TNT alone and TNT combined with pembrolizumab for LARC, and the initial results showed that the pCR of the two groups (regardless of MMR status) were similar (31.9% vs 29.4%, P = 0.75)[15]. Taken together, these studies show that only some patients with MSS CRC are responsive to immunotherapy. The NSABP FR-2 trial showed that LARC patients who received immunotherapy after preoperative radiotherapy and chemotherapy had a pCR of 22.2%, continuous clinical remission of 31.1%[23] and no significant increase in short-term adverse reactions. The ANAVA trial of patients with LARC showed that the MPR was 61.5% and the pCR was 21.8% when 6 cycles of avelumab immunotherapy were administered after radiotherapy and chemotherapy[24]. Other studies showed that LARC patients who had better immune function, increased antigen exposure, a longer immune memory before surgery, and receipt of radiotherapy and chemotherapy, benefitted the most from immunotherapy[22,33].

There have been recent reports on the application of novel neoadjuvant immunotherapy regimens for LARC patients in China. For example, the TORCH trial examined LARC patients, 80.2% of whom had a tumor less than 5 cm from the anal margin. The results confirmed that a regimen of short-course radiotherapy (SCRT) combined with chemotherapy and immunotherapy as a TNT improved the rate of tumor regression and anal preservation (82.3% to 86.4%)[8]. Another study involving Chinese patients with pMMR LARC, participants were administered either a control treatment (NACRT) or an experimental treatment (NACRT combined with neoadjuvant immunotherapy). The total CR was significantly greater in the experimental group (44.8% vs 26.9%, P = 0.031). Additionally, 37.3% of patients with a tumor diameter less than 5 cm had an anal preservation rate of 94.7%[9]. The PANDORA trial[17] found that the CCR of LARC patients after neoadjuvant immunotherapy combined with radiotherapy and chemotherapy was 25.4%. Patients in NRG-GI002 trial[15] had a CCR of 13.9%, and a significantly increased proportion of these patients adopted the approach of watchful waiting. The results of our meta-analysis demonstrated that neoadjuvant immunotherapy combined with radiotherapy and chemotherapy significantly improved the rate of anus-preservation in LARC patients.

Many recent randomized controlled studies have confirmed the advantages of TNT combined with short-term radiotherapy and long-term radiotherapy and chemotherapy for LARC, but the optimal mode and timing of these different treatments are still uncertain. There is evidence that whole-course neoadjuvant therapy can improve the effect of postoperative adjuvant chemotherapy, leads to better patient compliance and tolerance to chemotherapy than the standard regimen, and thus leads to improvements in tumor regression, pCR, organ preservation, and disease-free survival[4,34-37]. The NSABP FR-2 and PANDORA trials showed that the pCR of LARC patients to a new long-term synchronous radiotherapy and chemotherapy combined with immunotherapy was over 30%[17,23]. The NRG-GI002 trial[15] showed that TNT combined with immunotherapy improved the pCR rate of LARC patients compared with TNT alone (40% vs 29%), and that the adverse effects were manageable. The NECTAR trial[38] confirmed the safety and effectiveness of TNT combined with a PD-1 monoclonal antibody (tislelizumab for the treatment of MSS/pMMR LARC). The end-point data showed that the pCR was 40%, the ORR was 76.1%, and the organ retention rate was about 90%. The PKUCH 04 trial in Peking showed that TNT combined with neoadjuvant immunotherapy improved the pCR of patients with pMMR/MSS rectal cancer (33.3%)[19]. A single large dose of SCRT can increase tumor cell apoptosis and antigen release, and some researchers believe this approach can lead to a synergistic effect when combined with immunotherapy. Thus, many clinical studies of neoadjuvant radiotherapy combined with immunotherapy for LARC have mostly adopted the use of SCRT[39]. The TORCH, UNION, and AVERECTAL trials all reported positive results and excellent performance when using SCRT. In summary, whether using long-term synchronous radiotherapy and chemotherapy, neoadjuvant therapy, or SCRT, the addition of immunotherapy can provide significant advantages. At present, the combination of SCRT with immunotherapy seems to provide better short-term benefits. However, further research and long-term follow-up results are needed to identify the optimal treatment regimen.

The timing of neoadjuvant radiotherapy, chemotherapy, and ICI therapy may affect the outcome, and previous studies have examined the effects of three regimens: (1) Radiotherapy and chemotherapy after immunotherapy induction; (2) Synchronous immunotherapy with neoadjuvant radiotherapy and chemotherapy; and (3) Immune consolidation after sensitization by radiotherapy. The PKUCH 04 study[19] examined the effect of the first regimen with karelizumab for the treatment of LARC with MSS/pMMR. Thus, patients received three cycles of induction therapy (oxaliplatin + capecitabine + karelizumab), followed by long-term radiotherapy, and then two cycles of capecitabine consolidation therapy. All 25 eligible patients completed the preset treatment plan, 21 patients underwent surgery, the pCR was 33.3%, and 4 patients chose watchful waiting. The AVANA trial examined the effect of the second regimen[24]. These researchers administered six cycles of averuzumab simultaneously with neoadjuvant radiotherapy and chemotherapy, and recorded postoperative pCR. They obtained pCR values in 22 of 96 patients (23%) for whom a pathological response could be evaluated, and the incidence of grade 3 and above immune-related adverse reactions was only 4%. Rahma et al[15] also examined the effect of the second regimen. They added immunotherapy during radiotherapy and chemotherapy, followed by 8 cycles of mFOLFOX6; the control group received radiotherapy combined with oral capecitabine chemotherapy, followed by 8 cycles of mFOLFOX6. Their results showed that the two groups had similar rates of tumor regression, and that immunotherapy did not lead to adverse effects. The PANDORA trial examined the third regimen (immunoconsolidation after radiotherapy sensitization)[17]. Thus, patients with LARC received duvalizumab three times (once every four weeks) after long-term synchronous radiotherapy and chemotherapy. The pCR was 34.5% and there were no grade 4 toxic reactions, suggesting that the use of this regimen with duvalizumab can lead to a good pathological response and has satisfactory safety. The TORCH trial used a “pick the winner” approach and compared the results of an induction group (28 cases) and a consolidation group (34 cases). The induction group received two courses of chemotherapy + immunotherapy, followed by SCRT, and finally four courses of chemotherapy + immunotherapy; the consolidation group received SCRT first, and then six courses of CAPOX combined with toripalimab (anti-PD-1 monoclonal antibody). For the 32 patients who underwent TME, the induction group had a significantly better pCR (69.2% vs 47.4%) and CCR (55.9% vs 35.7%)[8]. Further studies are needed to determine the optimal timing of these combination therapies.

Our meta-analysis showed that hematological toxicity and abnormal liver function were the most common adverse events above grade 3. The adverse reactions to neoadjuvant radiotherapy and chemotherapy are mainly leukopenia and an increase of transaminase, and these are mostly mild and tolerable. The incidence of immune-related adverse reactions is low, although impaired endocrine function is common. Most of the immune-related adverse events were mild events (grade 1 or 2), and were treatable according to symptoms, although serious adverse events may require interruption of immunotherapy, hormone replacement therapy, and other treatments[40]. The main postoperative complications were infection, intestinal obstruction, and anastomotic stenosis. It is still uncertain whether neoadjuvant immunotherapy can lead to postoperative complications, a topic that requires further study.

CONCLUSION

The current meta-analysis has two main limitations. First, the heterogeneity among different outcomes were high. The I² ranged from 68% to 91%. Immunotherapy agents (PD-1 vs PD-L1), variations in chemoradiotherapy regimens (SCRT vs LCRT), differences in follow-up duration, and the proportion of patients with MSS/pMMR vs MSI/dMMR status. Second, very few publications have described the use of neoadjuvant therapy for LARC, and most of them were single-arm studies. Nonetheless, the results of our meta-analysis showed that neoadjuvant immunotherapy combined with radiotherapy and chemotherapy can improve the clinical remission rate (MPR, pCR, and CCR) of patients with LARC, especially for those with pMMR/MSS, and this treatment did not increase the incidence of postoperative complications and adverse reactions. We therefore suggest that neoadjuvant immunotherapy combined with radiotherapy and chemotherapy should be considered as a first choice for the treatment of LARC, including patients with pMMR/MSS. The results of ongoing phase III randomized controlled trials are needed for further confirmation.