Total Neoadjuvant Therapy and Organ Preservation in Locally Advanced Rectal Cancer: A State-of-the-Art Systematic Review and Critical Appraisal

1. Introduction

Rectal cancer represents a significant global health challenge. With colorectal cancer ranking as the third most common cancer and second leading cause of cancer mortality worldwide (1.9 million new cases, 904,000 deaths in 2022), rectal malignancies account for approximately one-third of this burden [1]. The management of locally advanced rectal cancer (LARC), encompassing stage II (T3-4, N0) and III (any T, N+) disease, has undergone a revolutionary transformation over the past three decades. The historical standard of surgery alone yielded unacceptably high rates of local recurrence and poor survival [2]. The sequential introduction of total mesorectal excision (TME), preoperative chemoradiotherapy (CRT), and adjuvant chemotherapy progressively improved outcomes, establishing a multimodal paradigm [3,4,5,6,7].

Despite these advances, distant metastasis persisted as the primary cause of death, highlighting the need for more effective systemic therapy. This led to the conceptualization and development of total neoadjuvant therapy (TNT), a strategy that administers all planned chemotherapy before surgery [8]. TNT aims to target micrometastatic disease earlier, improve treatment compliance, and enhance tumor downstaging [8,9]. Concurrently, the pivotal observation that patients achieving a clinical complete response (CCR) after neoadjuvant therapy could be safely managed with a non-operative “watch-and-wait” (WW) strategy established the foundation for organ preservation (OP), shifting a core therapeutic goal from mandatory radical resection to selective, response-adapted management [10].

Landmark phase III trials, including RAPIDO, PRODIGE 23, and STELLAR, have provided Level I evidence establishing the superiority of TNT over standard CRT for high-risk LARC [11,12,13]. The TNTCRT trial further confirmed the benefit of LCRT-based TNT with CAPOX, showing a 3-year DFS of 77.0% vs. 67.9% (HR 0.623, p=0.009) and a pCR rate of 27.5% vs. 9.9% [14]. Prospective studies like OPRA and OPERA have furnished robust Level II evidence supporting the oncologic safety and feasibility of OP [15,16,17]. We are now witnessing a second wave of innovation characterized by: (1) the successful integration of immunotherapy into TNT for MSS disease (UNION, STELLAR II, SPRING-01, PRECAM) [18,19,20,21,22]; (2) technological advances in radiotherapy delivery, such as proton therapy (PRORECT), MRI-guided radiotherapy (MRI-LINAC), and contact brachytherapy boosts (OPERA) [23,24]; and (3) the refinement of risk stratification through predictive biomarkers (Immunoscore biopsy, circulating tumor DNA (ctDNA) as in the CINTS-R trial and nuanced clinical response grading [25,26].

This rapidly evolving landscape presents new complexities and critical areas requiring further definition, including the optimization of chemotherapy and radiotherapy sequencing within TNT, the establishment of reliable criteria for CCR and the standardized grading of near-complete responses (NCR), the identification of ideal candidates for OP and the refinement of this selection through biomarkers, the delineation of the role for novel radiotherapy modalities in toxicity mitigation, and the development of tailored management strategies for challenging scenarios such as lateral pelvic lymph node involvement (LPLN+) and the growing population of elderly, potentially frail patients [9,27,28,29].

Furthermore, the optimal chemotherapy backbone within TNT remains debated. The MONT-R trial compared CapeOX versus capecitabine alone during neoadjuvant chemoradiotherapy in high-risk LARC, demonstrating that adding oxaliplatin significantly improved tumor regression (CAP 0-1: 58.6% vs. 46.8%; p=0.011) but did not translate into a 3-year DFS or OS benefit, with comparable pCR rates (25.5% vs. 25.3%). This suggests that while oxaliplatin enhances local tumor regression, its impact on long-term survival remains uncertain [30].

This systematic review aims to provide a comprehensive, contemporary synthesis of the evidence for TNT and OP in LARC. We move beyond reporting aggregate efficacy to dissect predictive factors for treatment response, analyze the impact of technological innovations, and offer evidence-based perspectives on managing specific clinical scenarios [9,31,32,33,34,35,36]. By integrating data from over 30 pivotal trials and their key ancillary analyses, this review serves as a definitive guide for clinicians navigating modern rectal cancer management and a roadmap for future translational and clinical research.

2. Materials and Methods

2.1. Search Strategy and Study Selection

A systematic literature search was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Electronic databases (PubMed, MEDLINE, Scopus, EMBASE) were searched from January 1990 through March 2026. Manual searches of conference abstracts from the European Society for Medical Oncology (ESMO) and the American Society of Clinical Oncology (ASCO) (2020–2026) were also performed. The search strategy employed a combination of Medical Subject Headings (MeSH) terms and free-text keywords, including: “rectal neoplasms”/”rectal cancer”, “neoadjuvant therapy”, “chemoradiotherapy”, “total neoadjuvant therapy”, “immunotherapy”, “PD-1 inhibitor”, “organ preservation”, “watch-and-wait”, “non-operative management”, “contact x-ray brachytherapy”, “proton therapy”, “MRI-LINAC”, “ctDNA” , “quality of life”, “aged”/”elderly”, “lymph nodes, pelvic”.

2.2. Study Selection and Analytical Approach

We included phase II/III randomized controlled trials and major prospective phase II studies evaluating neoadjuvant strategies in non-metastatic LARC. Given the significant clinical and methodological heterogeneity across studies, particularly in chemotherapy regimens, radiotherapy schedules, and the integration of novel agents, a formal quantitative meta-analysis was deemed inappropriate. This review is therefore a systematic narrative synthesis (a state-of-the-art review). While the literature search was systematic, the analysis is interpretive and thematic, aiming to synthesize evidence for clinical application rather than to produce a pooled quantitative estimate. The process is summarized in a PRISMA flow diagram (Figure 1).

2.3. Protocol Registration

Clinical trial number: Not applicable. Although this work follows a systematic search strategy, it is reported as a state-of-the-art expert review and was prospectively registered in PROSPERO for transparency. The study protocol was under registration number CRD420251252675.

2.4. Data Synthesis

Given the significant clinical and methodological heterogeneity across studies, particularly in the choice of chemotherapy regimens, radiotherapy protocols (short-course vs. long-course), sequencing (induction vs. consolidation), and the integration of novel agents (immunotherapy), a formal quantitative meta-analysis was deemed inappropriate. Therefore, a narrative synthesis was performed. Results are organized thematically by major treatment strategies and research questions, supported by structured summary tables to facilitate comparison and interpretation.

3. Results

3.1. Study Selection and Characteristics

The initial database search yielded 2,847 records. After removal of duplicates and screening of titles/abstracts, 112 full-text articles were assessed for eligibility. A total of 45 key publications, originating from 30 distinct RCTs or major prospective phase II studies, were included in the final qualitative synthesis. Table 1 summarizes the fundamental characteristics of these primary trials and their key ancillary analyses, encompassing landmark TNT trials, immunotherapy-TNT combinations (including SPRING-01 and PRECAM), OP studies, predictive factor analyses (including CINTS-R), de-escalation trials (GRECCAR 4, FOWARC, CONVERT, PROSPECT), and trials evaluating advanced radiotherapy modalities such as proton therapy and MRI-LINAC [21,22,25,26]

3.1.1. MRI-Based Risk Stratification: Defining Three Clinical Risk Groups

Magnetic resonance imaging (MRI) is the cornerstone of baseline staging and treatment planning in LARC. The MERCURY and MERCURY II criteria have established that certain MRI features predict higher risks of local recurrence, distant metastasis, and poor response to neoadjuvant therapy. Based on these features, patients can be classified into three risk groups to guide treatment intensity. 

High-Risk Features (Indicate Need for TNT Intensification)

•

Threatened or involved mesorectal fascia (MRF+): Tumour or lymph node within ≤1 mm of the mesorectal fascia. Associated with local failure rates >20% after standard CRT alone.

•

cT4 category:

-

T4a: tumour invades the visceral peritoneum (peritoneal reflection). High risk of peritoneal dissemination.

-

T4b: tumour directly invades adjacent organs (bladder, prostate, seminal vesicles, sacrum, vagina, pelvic sidewall). Very high risk of involved margins and local relapse.

•

Extramural vascular invasion (EMVI+): Presence of tumour within veins outside the muscularis propria (serpentine appearance, expansion of vessel). Strongly associated with synchronous and metachronous distant metastases.

•

Bulky N2 disease: Multiple regional lymph nodes (≥4) or large nodes (>10-15 mm) in the mesorectum. Indicates high tumour burden and risk of residual disease after CRT.

•

Lateral pelvic lymph node involvement (LPLN+): Nodes in the internal iliac, obturator, or external iliac chains. Size ≥7 mm short axis, or ≥5 mm with malignant features (irregular border, mixed signal intensity, loss of fatty hilum). These nodes are not adequately treated by standard TME and may require lateral lymph node dissection (LLND) after TNT.

Intermediate-Risk Features (Standard CRT or TNT May Be Considered)

• cT3 with clear MRF (>1 mm)
• cN1 (1-3 small nodes)
• No EMVI
• No LPLN enlargement

Low-Risk Features (De-Escalation Possible)

• cT2-3, MRF-, N0-1 (non-bulky), EMVI-, LPLN-
• Mid-upper rectal location (≥5-10 cm from anal verge)

Table 2 summarises these risk groups, the supporting MRI features, and the recommended treatment intensity based on current evidence.

Abbreviations: AV = anal verge; CRT = chemoradiotherapy; CT = chemotherapy; LLND = lateral lymph node dissection; LPLN = lateral pelvic lymph node; MRF = mesorectal fascia; TME = total mesorectal excision; TNT = total neoadjuvant therapy.

3.2. The Efficacy of Total Neoadjuvant Therapy: A Confirmed Benefit with a Critical Caveat

The efficacy outcomes of the three pivotal phase III trials are summarized in Table 3. As illustrated in Figure 2, these trials evaluated distinct TNT sequences, consolidation (RAPIDO, STELLAR) and induction (PRODIGE 23), all demonstrating that TNT significantly improves disease control compared to standard CRT. Our synthesis confirms that, for high-risk LARC, TNT represents a superior therapeutic strategy compared to traditional CRT followed by surgery and adjuvant chemotherapy (Level of evidence: I).

*Statistically significant (p < 0.05). Abbreviations: MFS = metastasis-free survival; MPR = major pathological response; NS = not significant; other abbreviations as in.

The PRODIGE 23 trial, utilizing an induction chemotherapy approach (induction FOLFIRINOX → LCRT → TME → mFOLFOX: Figure 2C), demonstrated a significant improvement in 7-year DFS (67.6% vs. 62.5%; HR 0.69) and, critically, a 7-year overall survival (OS) benefit (81.9% vs. 76.1%; HR 0.72), the first OS benefit in the TME era. The pCR rate was more than doubled (27.5% vs. 11.7%) [12,37].

The RAPIDO trial established that a short-course radiotherapy (SCRT)-based TNT regimen (SCRT → chemotherapy → TME: Figure 2B) significantly reduced 3-year disease-related treatment failure (23.7% vs. 30.4%; HR 0.75), primarily driven by a reduction in distant metastases (20.0% vs. 26.8%)[11]. pCR rates doubled (28% vs. 14%).

The STELLAR trial from an Asian population confirmed the non-inferiority of an SCRT-based TNT approach (SCRT → CAPOX → TME: Figure 2A) compared to LCRT followed by TME and adjuvant chemotherapy, with 3-year DFS rates of 64.5% vs. 62.3%, respectively [13].

The TNTCRT trial provided additional phase III evidence using LCRT-based TNT with CAPOX (one cycle induction, two cycles concurrent, three cycles consolidation). At a median follow-up of 44 months, the 3-year DFS was significantly improved in the TNT arm (77.0% vs. 67.9%; HR 0.623, 95% CI 0.435-0.892, p=0.009). The 3-year metastasis-free survival was also significantly higher (83.0% vs. 74.2%; HR 0.595, p=0.013). The pCR rate was 27.5% in the TNT arm vs. 9.9% in the standard arm (OR 3.436, p=0.0001). Importantly, this trial used LCRT (50-50.4 Gy) rather than SCRT, demonstrating that LCRT-based TNT is also highly effective [14].

These consistent findings across diverse geographical populations and TNT sequencing strategies solidify TNT as the standard of care for high-risk LARC. The primary oncologic benefit is enhanced systemic control, which translates into reduced distant metastases and, as evidenced by PRODIGE 23, an eventual survival advantage. The doubling or tripling of pCR rates is not merely a surrogate but a gateway to organ preservation [34,37].

However, a critical trade-off must be acknowledged and interrogated. The 5-year follow-up of the RAPIDO trial revealed a significantly higher incidence of locoregional recurrences in the experimental TNT arm compared to the standard arm (10% vs. 6%; p=0.027) [38]. This finding has become a central point of debate.

The long-term results of the Polish II trial provide important context. With a median follow-up of 7.0 years, this trial compared SCRT with three cycles of FOLFOX4 versus chemoradiation (50.4 Gy with bolus 5-FU, leucovorin, and oxaliplatin) in cT4 or fixed cT3 rectal cancer. The initial early OS benefit favoring SCRT-based TNT (9% at 3 years) disappeared with longer follow-up; at 8 years, OS was 49% in both groups (HR 0.90; 95% CI 0.70-1.15; p=0.38) [39]. No significant differences were observed in DFS, local failure (35% vs. 32%), or distant metastases (36% vs. 34%). Late complication rates were similar (grade 3+: 11% vs. 9%). These findings suggest that the early OS benefit observed with SCRT-based TNT may not be sustained, and the choice of radiotherapy platform requires careful consideration.

Why did RAPIDO potentially fail on local control? Several hypotheses warrant consideration: (1) Inadequate radiosensitization. The SCRT backbone (5x5 Gy) may provide insufficient biological dose intensity for bulky, locally advanced tumors compared to the protracted, radiosensitized course of LCRT; (2) prolonged interval to surgery. The long interval between radiotherapy and surgery in the consolidation TNT arm may allow for tumor cell repopulation in radioresistant clones; (3) Patient selection. The high-risk population in RAPIDO (e.g., cT4, N2, EMVI+, MRF+) may require the more potent local effects of LCRT; and (4) Loss of adjuvant chemotherapy effect.

This observation has led to nuanced clinical recommendations. ASCO 2024 guidelines issued a conditional preference for LCRT over SCRT within TNT regimens, particularly when maximizing local control is paramount [38]. ESMO 2025 guidelines consider both modalities valid but acknowledge LCRT may be preferred for very high-risk local tumors or when OP is the primary objective.

3.2.1. Baseline Heterogeneity Among TNT Trials: A Critical Comparison

A fundamental limitation of cross-trial comparisons of TNT efficacy is the marked heterogeneity in patient populations. Table 4 summarizes the key baseline risk factors for the major phase III TNT trials and selected immunotherapy/de-escalation studies.

As shown in Table 4, the RAPIDO trial enrolled the highest-risk population, with 74% cT4 tumors, 86% cN2, and 53% EMVI+. In contrast, PRODIGE 23 had only 26% cT4 (and a lower proportion of cN2, approximately 70%). STELLAR was intermediate with 37% cT4 and 74% cN2. The TNTCRT trial also enrolled a high-risk population (cT4, cN2, EMVI+, MRF+, or enlarged lateral nodes) and reported a pCR rate of 27.5% with LCRT-based TNT [14].

These differences have profound implications:

• The absolute pCR rate (28% in RAPIDO vs 27.5% in PRODIGE 23) appears similar, but the RAPIDO population was much higher risk. Therefore, the relative benefit of TNT over standard CRT may be greater in higher-risk patients.
• The increased locoregional recurrence seen in RAPIDO (10% vs 6%) may be partly explained by the high-risk features (cT4, EMVI+), which are known to predispose to local failure, and by the use of SCRT instead of LCRT.
• De-escalation trials (PROSPECT, CONVERT, FOWARC) intentionally excluded many high-risk patients (e.g., cT4, MRF+), so their results cannot be generalized to the high-risk LARC population.

Consequently, any comparison of outcomes across trials must account for these baseline differences. Clinicians should select TNT regimens based on the patient’s risk profile and the trial that best matches that profile, rather than assuming all TNT strategies are equally effective across all risk strata.

3.3. De-Escalation: The PROSPECT Trial and Beyond

Parallel to TNT intensification, the PROSPECT trial established a valid de-escalation option for selected lower-risk patients [40]. In patients with mid-to-upper rectal tumors (cT2 N1, cT3 N0/N1, CRM-negative), neoadjuvant FOLFOX (6 cycles) with selective salvage LCRT for poor response (required in only 9.1%) was non-inferior to standard LCRT for 5-year DFS (80.8% vs. 81.7%). This strategy spares patients the long-term toxicity of pelvic radiation (particularly sexual dysfunction) at the cost of increased acute chemotherapy-related toxicity.

The CONVERT trial further explored radiotherapy omission in LARC with uninvolved mesorectal fascia (MRF-). In this phase III trial of 663 patients, neoadjuvant CAPOX alone (4 cycles) was compared to capecitabine-based nCRT. The non-inferiority of nCT was not formally confirmed due to a very low incidence of local recurrence in both groups (3-year LRRFS: 96.3% for nCT vs. 97.4% for nCRT; HR 1.40, 95% CI 0.53-3.68). However, nCT offered comparable DFS (89.2% vs. 87.9%) and OS (95.0% vs. 94.1%) while significantly reducing grade 2-4 long-term adverse events (16.0% vs. 26.3%, p=0.002) and proctitis (33.6% vs. 41.7%, p=0.049). These findings support radiotherapy omission in carefully selected patients with uninvolved MRF [41].

The FOWARC trial long-term results with a median follow-up of 10 years compared mFOLFOX6 with or without radiation versus fluorouracil plus radiation. The 10-year DFS rates were 52.5%, 62.6%, and 60.5%, respectively (p=0.56), and 10-year LR rates were 10.8%, 8.0%, and 9.6% (p=0.57). Patients achieving pCR had excellent outcomes (10-year DFS 84.3%, OS 92.4%). These long-term data confirm that neoadjuvant mFOLFOX6 alone can be a therapeutic option for LARC without compromising local control or survival [7].

The GRECCAR 4 trial evaluated a response-adaptive strategy: patients with LARC received induction chemotherapy, and good responders proceeded directly to surgery without CRT, while poor responders received CRT. This approach demonstrated the feasibility of tailoring treatment intensity based on early response, avoiding unnecessary radiotherapy in good responders [42].

These de-escalation and response-adaptive approaches demonstrate that a “one-size-fits-all” approach is obsolete; the availability of both intensification (TNT) and de-escalation (PROSPECT, CONVERT, FOWARC, GRECCAR 4) enables truly personalized treatment.

3.4. Immunotherapy in MSS Disease: A Paradigm Shift in Waiting, Not Yet Realized

A transformative, but still investigational, development is the integration of immune checkpoint inhibitors (ICIs) into TNT for MSS rectal cancer. The UNION trial (camrelizumab) [32] and STELLAR II trial (sintilimab) [47] have reported remarkable increases in pCR and clinical CR rates, approximately doubling them compared to TNT alone (e.g., UNION: 39.8% vs. 15.3%) (Level of evidence: II, pending mature phase III confirmation) [20,43,44]. The Averectal study (avelumab) reported a pCR rate of 36% [45].

The SPRING-01 trial, a randomized phase II trial, compared SCRT followed by sintilimab plus CAPOX versus SCRT followed by CAPOX alone in 98 patients with LARC. The pCR rate was significantly higher in the immunotherapy-containing arm: 59.2% (95% CI 45.4-72.9) vs. 32.7% (95% CI 19.5-45.8) (p=0.015). The complete response rate (pCR + cCR) was also significantly improved (61.2% vs. 32.7%; OR 3.2, 95% CI 1.4-7.5, p=0.0085). Grade 3-4 treatment-related adverse events occurred in 33% vs. 35%, with no treatment-related deaths in the immunotherapy arm. These results are among the highest pCR rates reported in MSS LARC to date [46].

The PRECAM study evaluated short-course nCRT (25Gy/5f) followed by two cycles of CAPEOX and six weekly doses of enzalofilimab (a PD-L1 antibody) in 34 patients with MSS LARC. The pCR rate was 62.5% (20/32), and the major pathologic response rate (TRG 0-1) was 75%. Common adverse events were manageable, with only two grade 3 events (liver function abnormality and thrombocytopenia). These findings further support the potent synergy between short-course radiotherapy and PD-1/PD-L1 blockade [22,47].

The NRG-GI002 trial (long-term results) evaluated pembrolizumab added to TNT in a phase II platform. With longer follow-up, the addition of pembrolizumab was associated with a statistically significant improvement in 3-year OS (95% vs. 87%; HR 0.35, 95% CI 0.12-1.00, p=0.04), but not DFS (64% vs. 64%; HR 0.95, p=0.82). The neoadjuvant rectal (NAR) score improvement was not statistically significant (mean difference 2.9, p=0.21). These results suggest a potential OS benefit that requires confirmation in larger trials [48].

While these results are highly promising and suggest that radiotherapy can act as an in-situ vaccine to overcome immune evasion in MSS disease, a cautious interpretation is essential for several reasons: (1) Phase II data predominates: Most of these results are from phase II trials or early analyses of phase III trials with small sample sizes. Mature survival data (DFS, OS) are not yet available for many; (2) Heterogeneous designs: In the UNION trial, the control and experimental arms differed in both systemic therapy and radiotherapy platform; (3) Toxicity signals: The addition of ICIs increases the risk of immune-related adverse events (irAEs); and (4) Variability in pCR rates: pCR rates range from 36% (Averectal) to 62.5% (PRECAM), indicating that optimal regimens, sequencing, and patient selection are not yet defined.

Therefore, these results must be considered hypothesis-generating. While they provide a powerful signal and justify rapid further investigation, the addition of immunotherapy to TNT for MSS rectal cancer cannot currently be recommended for routine clinical implementation outside the context of a clinical trial.

3.5. Organ Preservation: Proven Efficacy in Expert Centers, Questions of Generalizability

The high pCR rates achieved with TNT have legitimized OP as a viable oncologic strategy. The randomized phase II OPRA trial provided pivotal evidence, demonstrating that a WW approach for good responders yields long-term OP in over 50% of patients without compromising survival [15,49]. Its updated 5-year results are central to clinical practice: (1) TME-free survival: Significantly higher with the consolidation chemotherapy sequence (54% vs. 39% for induction), supporting consolidation as the preferred sequence when OP is a goal; (2) Safety of Salvage: 94% of tumor regrowths occurred within the first 2 years, and salvage TME for regrowth resulted in equivalent 5-year DFS (64%) as immediate TME for incomplete responders; and (3) Predictive response grading: The validation of a three-tier clinical response system (CCR, NCR, Incomplete Response/ICR) provides a powerful prognostic tool [49]. Patients with a sustained CCR had a 3-year OP rate of 77% and a 3-year DFS of 88%, whereas those with an NCR had rates of 40% and 69%, respectively.

The CAO/ARO/AIO-16 trial provided additional prospective data on OP after TNT, reporting a CCR rate of 36% and a 3-year OP rate of 68% in patients achieving CCR. Patients with sustained CCR had significantly better bowel function (lower LARS/Wexner scores) compared to those undergoing immediate TME [50].

The MONT-R trial also included an ancillary study evaluating transanal endoscopic microsurgery (TEM) versus radical surgery in patients achieving cCR or near-cCR after nCRT. At a median follow-up of 60 months, TEM was associated with significantly faster recovery, better sphincter function (Wexner: 1 vs. 4, p=0.001; LARS: 0 vs. 17, p<0.001), and comparable 5-year DFS (75.6% vs. 80.9%, p=0.658) and OS (93.2% vs. 88.2%, p=0.465). These findings support local excision as an alternative to TME for selected good responders, preserving function without compromising oncologic outcomes [30,49].

3.6. Predictive Factors and Risk Stratification: Informing Personalization and the Role of ctDNA

Identifying predictors of response is crucial for refining patient selection. A post-hoc analysis of the RAPIDO trial identified that the use of TNT itself (OR 2.70), a pretreatment CEA level <5 µg/L, and a tumor size <40 mm were independent predictors of achieving a pCR [38]. Achieving pCR was associated with excellent prognosis (5-year OS >90%).

Conversely, the presence of LPLN+ remains a stubbornly negative prognostic factor, even in the TNT era. A subgroup analysis from the STELLAR trial showed that patients with LPLN+ had inferior 3-year DFS compared to those without (51.7% vs. 66.2%) [51].

The CINTS-R trial represents a major advance in precision medicine for LARC. This multicenter randomized controlled trial uses ctDNA to guide neoadjuvant treatment intensity. In the experimental group, patients with high baseline ctDNA abundance (median VAF ≥0.5%) or persistent ctDNA positivity after CRT receive TNT (nCRT plus 4-6 cycles of XELOX), while ctDNA-low-risk patients receive conventional nCRT. Patients with dMMR/MSI-H or TMB-H receive neoadjuvant immunotherapy. The interim analysis of 349 patients demonstrated feasibility and safety: serious adverse events occurred in 10.0% of the experimental group vs. 6.6% of the control group (p=0.316). Notably, 15.7% of TNT-treated patients discontinued chemotherapy due to SAEs, whereas all nCRT recipients completed treatment. Importantly, traditional clinical risk stratification and ctDNA-guided stratification showed substantial discordance: 43.0% of traditionally high-risk patients were classified as ctDNA-low-risk, and 53.1% of traditionally low-risk patients were classified as ctDNA-high-risk. This highlights the potential of ctDNA to refine patient selection beyond clinical factors alone. The primary endpoint (2-year DrTF rate) is awaited [25,26].

3.7. Toxicity, Compliance, and Patient-Reported Quality of Life

As summarized in Table 5, the intensified nature of TNT regimens results in a higher incidence of acute grade ≥3 toxicities during the neoadjuvant phase, primarily hematological (e.g., neutropenia) and gastrointestinal (e.g., diarrhea), compared to standard LCRT alone. Rates ranged from 26.5% in STELLAR to 47.6% in RAPIDO for the TNT arms [3,29] [11,13]. The MONT-R trial reported grade 3-4 toxicity rates of 14.1% with CapeOX vs. 9.3% with capecitabine alone (p=0.095) [30,51]. Compliance with the planned preoperative therapy was generally high (>80% completion).

However, comprehensive longitudinal quality of life (QoL) assessments from these trials provide a reassuring and nuanced picture. The RAPIDO QoL study found no significant differences in global health status, bowel function (assessed by the Low Anterior Resection Syndrome score), or late grade ≥3 toxicity at 3 years post-surgery between TNT and standard care groups [52]. The PRODIGE 23 analysis revealed that while neoadjuvant FOLFIRINOX transiently reduced global QoL, scores recovered by 2 years and converged with the standard arm [53]. The STELLAR trial reported no clinically significant difference in global QoL or anal function at 6 years [54]. Additionally, a dosimetric comparison from the PRORECT trial suggested proton therapy may significantly reduce the risk of acute GI toxicity [24].

The CONVERT trial demonstrated that nCT alone significantly reduced grade 2-4 long-term AEs (16.0% vs. 26.3%, p=0.002) and proctitis (33.6% vs. 41.7%, p=0.049) compared to nCRT, with comparable DFS and OS [41].

The Polish II trial reported no significant differences in late complication rates between SCRT-based TNT and chemoradiation (grade 3+: 11% vs. 9%, p=0.66) [39].

While TNT imposes a higher acute treatment burden, it does not appear to inflict permanent detriment to long-term QoL for most patients. The decision to pursue TNT involves balancing this transient burden against the long-term benefits of reduced metastasis risk, survival gain, and the chance for organ preservation.

3.8. Technological Advances in Radiotherapy: MRI-LINAC and Adaptive Planning

Beyond proton therapy, magnetic resonance-guided radiotherapy (MRI-LINAC) represents a transformative technological leap. Chen et al. developed an efficient library of reference plans (LoRP) strategy for MRI-guided adaptive radiotherapy of rectal cancer [55]. This approach involves preparing multiple reference plans based on diverse bladder shapes; for each fraction, a plan is selected based on daily bladder filling. Compared to fully adaptive (adapt-to-shape) strategies, LoRP reduced treatment session duration by more than a third (>20 minutes) while maintaining acceptable target coverage (94% vs. 95% for fully adaptive, and 92% of LoRP plans achieved acceptable dose criteria vs. 74% for conventional couch shift). This strategy enhances treatment efficiency, patient comfort, and enables real-time adaptation to anatomical changes.

3.9. Health Economics and Real-World Feasibility

The adoption of TNT, particularly with novel agents and advanced technologies, raises significant questions of sustainability and equity. The cumulative cost of intensified chemotherapy, ICIs (when used), CXB boosts, and intensive surveillance protocols for WW is substantial. Rigorous cost-effectiveness analyses are urgently needed to determine the value of these strategies across different healthcare systems. In low- and middle-income countries, the routine implementation of such resource-intensive protocols is currently unfeasible. Therefore, future research and guideline development must consider not only efficacy but also affordability and accessibility, ensuring that advances in rectal cancer care do not exacerbate global health inequities.

3.10. Synthesis of Key Findings

The integrated analysis of the included trials yields several key conclusions that collectively inform the current management paradigm for locally advanced rectal cancer (Figure 3).

First, total neoadjuvant therapy improves systemic control for high-risk LARC. The RAPIDO and PRODIGE 23 trials provide Level I evidence that TNT reduces distant metastases and improves disease-related treatment failure or DFS compared to standard CRT, with PRODIGE 23 now demonstrating an overall survival benefit [11,12,37]. The STELLAR trial supports the non-inferiority of an SCRT-based TNT regimen for 3-year DFS compared to LCRT-based therapy [13]. The TNTCRT trial provides additional phase III evidence for LCRT-based TNT with CAPOX [14].

Second, TNT carries an inherent trade-off between systemic and local control. While improving distant metastasis rates, some TNT regimens, particularly the SCRT-based consolidation approach used in RAPIDO, are associated with a higher risk of locoregional recurrence compared to standard chemoradiotherapy. The Polish II trial showed that early OS benefits of SCRT-based TNT were not sustained with longer follow-up, with 8-year OS of 49% in both arms [39].

Third, immunotherapy shows highly promising activity in MSS disease, but these results remain investigational. The UNION, STELLAR II, SPRING-01, and PRECAM trials demonstrate that adding a PD-1/PD-L1 inhibitor to SCRT-based TNT markedly increases pCR and cCR rates (approximately 40-60% vs. 15-25%) in MSS rectal cancer [18,20]. SPRING-01 reported the highest pCR rate (59.2%) in a randomized trial to date. However, these findings are primarily from phase II or early-phase III studies with heterogeneous designs and without mature survival data.

Fourth, organ preservation is confirmed as a safe and viable oncologic strategy in expert centers. Prospective trials including OPRA, OPERA, and CAO/ARO/AIO-16 demonstrate that non-operative management for good responders yields long-term OP in over 50% of selected patients without compromising survival [15,16,50]. The MONT-R TEM study further supports local excision as an alternative to TME for good responders [30,51].

Fifth, therapeutic sequence significantly impacts OP rates. The consolidation chemotherapy sequence yields higher rates of both pCR and OP compared to the induction sequence [15,56].

Sixth, short-course radiotherapy is validated as an efficient TNT backbone, but its use requires careful consideration of local control trade-offs. The STELLAR and Polish II trials confirm that SCRT followed by chemotherapy is non-inferior to conventional long-course radiotherapy for long-term survival, offering a shorter and more resource-efficient treatment platform [13,35,52,57]. However, the Polish II long-term results showed no sustained OS benefit.

Seventh, predictive factors and refined response grading significantly advance personalization. The CINTS-R trial is pioneering ctDNA-guided risk stratification, with interim analysis confirming feasibility and safety [25,26]. The OPRA three-tier clinical response system provides a powerful prognostic tool [20,49].

Eighth, advanced radiotherapy modalities can mitigate treatment toxicity. Dosimetric data from the PRORECT trial suggest that proton therapy could significantly reduce acute gastrointestinal toxicity [23]. MRI-LINAC with library of reference plans enables efficient adaptive radiotherapy, reducing treatment session duration while maintaining target coverage [56].

Ninth, de-escalation and response-adaptive strategies are feasible in selected patients. The PROSPECT, CONVERT, FOWARC, and GRECCAR 4 trials demonstrate that radiotherapy can be safely omitted or adapted based on response in lower-risk LARC, reducing long-term toxicity without compromising survival.

Tenth, specific clinical subgroups demand tailored approaches. LPLN+ remains a negative prognostic factor despite TNT, warranting consideration of treatment intensification [52]. Elderly patients require individualized care focused on function preservation, often leveraging the efficacy of radiotherapy and organ preservation strategies guided by Comprehensive Geriatric Assessment [9,27].

Eleventh, real-world evidence supports feasibility and tolerability. Large cohort studies, such as the Swedish nationwide LARCT-US project, confirm that SCRT-based TNT (with a modified, shorter chemotherapy course) is feasible, effective, and associated with manageable toxicity in routine clinical practice [17].

Finally, long-term QoL is preserved with TNT, as despite higher acute toxicity, it does not impair long-term global quality of life, bowel, or anal function compared to standard CRT [52,53,54].

4. Discussion

4.1. Implementation of Total Neoadjuvant Therapy

For patients with high-risk LARC, TNT is now a preferred strategy supported by landmark phase III trials [11,12,13]. By delivering all systemic chemotherapy preoperatively, TNT targets micrometastatic disease earlier, improves treatment compliance, and achieves superior tumor downstaging, tripling pCR rates and creating opportunities for organ preservation [11,12,13,56].

However, the critical trade-off revealed by the RAPIDO trial, improved systemic control at the cost of increased locoregional recurrence (10% vs. 6%), must centrally inform treatment decisions [11,38]. The Polish II trial long-term results further caution that early OS benefits may not be sustained. This observation reframes TNT not as a monolithic regimen but as a flexible framework requiring deliberate optimization of two sequential choices.

The first decision is the radiotherapy platform. The STELLAR trial validated SCRT-based TNT as non-inferior to LCRT for 3-year DFS, offering efficiency and resource savings [13]. Real-world data from the LARCT-US cohort further support its feasibility [17]. However, in light of RAPIDO’s local failure signal and the Polish II long-term results, LCRT may be preferred for tumors at highest risk of local recurrence (e.g., low-lying tumors, MRF+), where protracted radiosensitization may be advantageous. The TNTCRT trial provides strong evidence for LCRT-based TNT with CAPOX, demonstrating significant DFS and MFS benefits [14].

The second decision is treatment sequencing. Induction chemotherapy (PRODIGE 23) facilitates rapid cytoreduction for symptomatic or bulky tumors [12]. Consolidation chemotherapy (RAPIDO, OPRA) maximizes the likelihood of deep clinical response and is therefore preferred when organ preservation is the primary [11,15].

The optimal chemotherapy backbone also requires consideration. The MONT-R trial demonstrated that adding oxaliplatin to capecitabine during nCRT improves tumor regression but does not translate into DFS or OS benefits, suggesting that the intensification of local therapy does not always improve systemic outcomes [30].

Thus, TNT application requires individualization: selecting the appropriate radiotherapy backbone based on local risk, then the optimal sequence based on treatment objectives, always balancing the systemic-local control trade-off (Figure 3).

Crucially, the marked heterogeneity in baseline risk factors across TNT trials must be acknowledged when interpreting outcomes (Table 4). RAPIDO enrolled a much higher-risk population (74% cT4, 53% EMVI+) than PRODIGE 23 (26% cT4). Despite this, both trials achieved similar pCR rates (approximately 28%). This suggests that the relative benefit of TNT over standard CRT may be greater in higher-risk patients, but also that the trade-off with locoregional control is more pronounced in such patients. Therefore, a regimen that works well in a lower-risk population (e.g., PRODIGE 23) cannot be assumed to have the same safety profile in a very high-risk population. Clinicians should match the TNT regimen to the patient’s risk profile, considering both the radiotherapy platform (SCRT vs LCRT) and the intensity of chemotherapy.

4.2. Organ Preservation Paradigm and Precision Patient Selection

Organ preservation has become a standard option within specialized multidisciplinary care, redefining success to include functional integrity without compromising oncologic outcomes for selected patients.

The three-tier response grading system validated by the OPRA trial (CCR/NCR/ICR) is the cornerstone of precision patient selection [49]. This framework moves beyond the binary “CCR versus not” to enable risk-adapted management: (1) Sustained CCR: WW can be offered with high confidence; (2) NCR: WW remains feasible but requires intensive surveillance; and (3) ICR: Prompt TME is indicated.

The MONT-R TEM study provides strong evidence that local excision (transanal endoscopic microsurgery) is a safe alternative to TME for selected good responders, offering better functional outcomes (Wexner: 1 vs. 4, LARS: 0 vs. 17) with comparable 5-year DFS (75.6% vs. 80.9%) and OS (93.2% vs. 88.2%) [30].

Technical innovations further expand OP boundaries. The OPERA trial demonstrated that a CXB boost after CRT significantly improves OP rates (81% vs. 66% at 3 years) for small, distal tumors [16].

A critical caveat regarding generalizability must be emphasized. The outstanding outcomes from OPRA and OPERA were achieved in highly selected patients managed within high-volume expert centers. These results may not be immediately generalizable to all clinical settings.

4.2.1. Integrating MRI into Clinical Decision-Making

The use of high-quality baseline pelvic MRI is mandatory for all patients with LARC. Beyond simple TNM staging, MRI identifies key biological risk factors that dictate treatment intensity: (1) MRF+ predicts local failure even after good systemic therapy; such patients require TNT with careful attention to the radiotherapy platform (LCRT may be preferred over SCRT); (2) EMVI+ is a powerful predictor of distant metastasis; these patients derive the greatest benefit from the intensified systemic therapy of TNT (either induction or consolidation); (3) LPLN+ cannot be managed by standard TME alone. After TNT, post-treatment MRI should reassess lateral nodes: if they persist ≥7 mm or show malignant features, lateral lymph node dissection (LLND) should be considered (Ogura criteria); and (4) cT4b with organ invasion requires TNT and a multidisciplinary surgical plan (en-bloc resection may be necessary).

The three-risk-group framework (Table 4) allows clinicians to move beyond a one-size-fits-all approach. For high-risk patients, TNT is the preferred strategy, and participation in clinical trials of immunotherapy-TNT combinations is encouraged. For intermediate-risk patients, either standard CRT or TNT is acceptable, with the choice influenced by the patient’s desire for organ preservation (consolidation TNT preferred). For low-risk patients, de-escalation strategies (chemotherapy alone, selective CRT, response-adaptive care) are safe and reduce long-term toxicity.

MRI is also essential for post-TNT response assessment (mrTRG, residual nodal status, EMVI clearance). The combination of endoscopic (CCR/NCR/ICR) and MRI findings provides the most accurate prediction of pathological complete response and guides the watch-and-wait or local excision decision.

4.3. Immunotherapy Integration in MSS Tumors: Promise Requiring Confirmation

The integration of PD-1 inhibitors with TNT for MSS rectal cancer represents a highly promising but still investigational strategy [18,20]. The SPRING-01 trial reported a pCR rate of 59.2% with sintilimab-containing TNT, and the PRECAM study reported a pCR rate of 62.5% with enzalofilimab. These are among the highest pCR rates ever reported in MSS LARC [21,22].

However, a rigorously cautious interpretation is warranted for several reasons: (1) Maturity of Data: These results are predominantly from phase II trials. Mature survival data (DFS, OS) are not yet available; (2) Heterogeneous Designs: Different radiotherapy platforms (SCRT vs. LCRT), chemotherapy backbones, ICI agents, and sequencing are being used; (3) Toxicity Considerations: The addition of ICIs introduces irAEs; (4) Cost and Access: These regimens are substantially more expensive; and (5) Variability in pCR rates (36-62.5%) indicates that the optimal regimen is not yet defined.

Therefore, while these results are exceptionally promising, the addition of immunotherapy to TNT for MSS rectal cancer must currently be considered investigational outside of a clinical trial context.

4.4. Predictive Biomarkers and ctDNA-Guided Therapy: The Future of Personalization

Accurate response assessment is the cornerstone of organ preservation. While the OPRA trial’s three-tier clinical response grading system represents a major advancement, the inherent challenge persists that a CCR does not always equate to a pCR [49].

The CINTS-R trial represents a paradigm shift toward ctDNA-guided precision neoadjuvant therapy. By using baseline ctDNA abundance and post-CRT ctDNA clearance to stratify patients into TNT versus conventional nCRT (or immunotherapy for dMMR/TMB-H), this trial addresses the critical need for dynamic, biology-driven treatment adaptation. The interim analysis confirms feasibility and safety, and the final results (2-year DrTF rate) are eagerly awaited. If positive, ctDNA-guided therapy could become the new standard for personalizing neoadjuvant treatment intensity [25,26].

The Immunoscore biopsy, which quantifies immune cell densities within the tumor microenvironment, has also been validated as a predictor of local regrowth risk [59].

These sophisticated tools must be integrated with established clinicopathological factors, including pre-treatment CEA <5 µg/L and tumor size <40 mm [38].

4.5. Technological Advances in Radiotherapy: MRI-LINAC

MRI-guided radiotherapy (MRI-LINAC) represents a transformative technological leap. The library of reference plans (LoRP) strategy developed by Chen et al. addresses a key limitation of MRI-LINAC, the time required for full adaptive replanning. By preparing multiple reference plans based on varying bladder shapes and selecting the most appropriate plan for daily anatomy, LoRP reduced treatment session duration by >20 minutes compared to fully adaptive strategies, while maintaining acceptable target coverage (94% vs. 95%). This enhances patient comfort, reduces intrafraction motion, and improves the feasibility of daily adaptation in routine clinical practice [56].

4.6. Management of Specific Subgroups

Two persistent and complex clinical scenarios demand focused, nuanced strategies.

First, LPLN+ remains a significant negative prognostic factor even in the context of potent TNT, as evidenced by subgroup analyses from STELLAR [52]. The role and ideal timing of surgical lateral lymph node dissection in patients who have received TNT must be highly individualized, guided by meticulous multidisciplinary review of post-treatment MRI.

Second, the management of elderly and potentially frail patients requires a paradigm distinct from chronological age alone. A Comprehensive Geriatric Assessment is mandatory to objectively distinguish fit from frail individuals [9,27]. For fit older adults, curative-intent TNT remains appropriate. For the frail, the primary therapeutic aim shifts towards optimizing function and quality of life, with strategies including CRT with OP intent, definitive radiotherapy alone, or CRT followed by local excision.

4.7. Health Economics and Real-World Feasibility

The adoption of TNT, especially with new agents and advanced technologies, raises important questions about sustainability and equity. The total cost of intensified chemotherapy, immunotherapies, proton therapy, CXB boosts, and intensive surveillance protocols is high. In low- and middle-income countries, implementing such resource-heavy protocols is currently not feasible. Rigorous cost-effectiveness analyses are urgently needed.

4.8. Clinical Implications and Future Directions

Based on this comprehensive review, managing LARC should now be guided by evidence-based, yet nuanced, recommendations (Figure 3).

For patients with high-risk LARC (cT4, cN2, EMVI+, MRF+), TNT is a preferred approach to maximize systemic control [11,12,13].

The choice between SCRT and LCRT within TNT should balance systemic benefits against local control, with LCRT possibly favored for tumors at the highest risk of local failure. LCRT-based TNT with CAPOX (as in TNTCRT) is a valid and effective option [14].

When OP is a treatment goal, the consolidation sequence is preferred [13,15]. Response should be evaluated 8-14 weeks after completing TNT using a structured three-tier clinical response system (CCR, NCR, ICR) [20,49]. A WW approach should be offered to patients who achieve a sustained CCR and can be cautiously considered for selected patients with NCR, provided this is done within a strict, protocolized surveillance pathway in a specialized multidisciplinary center [15,49,53]. Local excision (TEM) is a safe alternative to TME for selected good responders (MONT-R TEM study) [30].

Regarding immunotherapy, adding a PD-1/PD-L1 inhibitor to TNT appears promising for increasing CCR rates in MSS patients [18,20]; however, this approach should currently be considered investigational and limited to clinical trials. The high pCR rates reported in SPRING-01 and PRECAM are encouraging but require phase III confirmation [21,22].

For patients with lower-risk LARC (e.g., MRF-, cT3N+ without high-risk features), de-escalation strategies including radiotherapy omission (PROSPECT, CONVERT, FOWARC) or response-adaptive approaches (GRECCAR 4) are valid options that reduce long-term toxicity.

ctDNA-guided therapy (CINTS-R) represents a promising future direction for personalizing neoadjuvant treatment intensity, but remains investigational pending final results.

For vulnerable populations, management must be carefully tailored: a CGA is essential for elderly patients [27]; and for those with LPLN+, treatment should involve intensified TNT with personalized decisions regarding lateral lymph node dissection [52].

The path forward in research and clinical practice will continue to evolve, emphasizing not only improved outcomes but also broader access and equity in care, and will prioritize several key areas:

• Immunotherapy confirmation: Phase III confirmatory trials with long-term survival data are urgently needed before immunotherapy-TNT combinations can be adopted as standard of care [18,20].
• ctDNA-guided therapy: The final results of the CINTS-R trial (2-year DrTF rate) will determine whether ctDNA-guided risk stratification should become standard practice.
• Biomarker integration: Validation and clinical implementation of Immunoscore, ctDNA dynamics, and radiomic signatures are paramount.
• Understanding and mitigating trade-offs: Further research is required to elucidate the mechanisms behind increased locoregional recurrence risk and optimize radiotherapy techniques.
• Health economic analysis: Rigorous cost-effectiveness analyses of TNT, OP strategies, and novel technologies are essential.
• Trials for specific subgroups: Dedicated prospective trials are required for elderly/frail patients and those with LPLN+.
• Radiotherapy advancement: Continued optimization of techniques, including proton therapy [23].
• MRI-guided adaptive radiotherapy, and CXB boosts [22], is crucial [16,23].

5. Conclusions

The management of LARC has entered an era of remarkable sophistication, driven by the conclusive evidence for TNT. This strategy has contributed to partially uncoupling survival from mandatory radical surgery, establishing a new standard of care for high-risk disease by improving systemic control and, as evidenced by PRODIGE 23 and TNTCRT, delivering DFS and OS benefits. Organ preservation is a validated and durable outcome for selected patients managed in expert centers, with local excision (TEM) emerging as a safe alternative to TME for good responders (MONT-R).

However, this review has deliberately highlighted the nuances and trade-offs that accompany this progress. The increased locoregional recurrence risk in the RAPIDO trial and the lack of sustained OS benefit in the Polish II trial serve as critical counterbalances. The dramatic pCR rates from immunotherapy-TNT combinations (SPRING-01: 59.2%; PRECAM: 62.5%), while paradigm-shifting in potential, remain investigational and require rigorous confirmatory trials. The MONT-R trial reminds us that adding oxaliplatin to nCRT improves tumor regression but not survival, highlighting the need to distinguish between surrogate endpoints and true clinical benefit.

The CINTS-R trial represents the future of precision medicine in LARC, using ctDNA to dynamically guide treatment intensity. The final results of this trial will be pivotal in determining whether biomarker-driven personalization can improve outcomes and reduce overtreatment.

We now possess more potent therapeutic tools and more refined lenses for assessment. The future of LARC care lies in the seamless synthesis of these elements: selecting the most effective yet best-tolerated strategy based on dynamic biomarkers (ctDNA, Immunoscore), advanced imaging (MRI, MRI-LINAC), and comprehensive patient assessment, with the unwavering dual objectives of achieving cure and preserving quality of life. This review charts the evidence-based, yet cautious, path toward that future of precision oncology, acknowledging both the remarkable progress made and the critical questions that remain.

“The future of rectal cancer management no longer lies in choosing between surgery and non-surgery, but in dynamically integrating systemic therapy, radiotherapy, biomarkers, and response assessment to deliver cure with maximal functional preservation.”