Factors associated with complete clinical response after neoadjuvant therapy in locally advanced rectal cancer: A Mexican retrospective study

Abstract

BACKGROUND

Treatment for rectal cancer varies widely among populations due to differences in tumor biology, access to treatment and care systems. Global responses to standard therapies are heterogeneous and data from Latin American populations, particularly Mexican patients, are scarce.

AIM

To assess complete clinical response (cCR) rates, local recurrence, and associated factors in a Mexican neoadjuvant therapy cohort.

METHODS

A retrospective observational study included 101 clinical stage II-III rectal cancer patients with neoadjuvant therapy. The primary endpoint was cCR, and pathological complete response (pCR) was assessed among patients who underwent surgery. An overall complete response rate (cCR or pCR) is reported descriptively. An exploratory univariate analysis was performed to evaluate factors associated with cCR. Response durability and local recurrence-free survival were estimated using the Kaplan-Meier method.

RESULTS

The cCR was achieved in 14.9% (n = 15) of the cohort, while overall complete response (cCR or pCR) was observed in 19.8% (n = 20). Among complete responders, 25% (n = 5) had pCR confirmed after surgery. During follow-up, two local recurrences occurred among patients managed with watch-and-wait (2/15; 13.3%), both within the first 16 months. In exploratory univariate analysis, tumor size was inversely associated with the likelihood of achieving cCR (OR = 0.77 per centimeter; 95%CI: 0.61-0.96; P = 0.028).

CONCLUSION

Neoadjuvant therapy achieved 19.8% of cCR, mostly managed by watch-and-wait. Tumor size predicted cCR. These findings support structured surveillance and warrant prospective studies for better predictors.

Key Words: Rectal neoplasms; Neoadjuvant therapy; Clinical complete response; Pathologic complete response; Watch-and-wait

Core Tip: Recognizing the high prevalence of locally advanced rectal cancer at presentation in the Mexican population, we assessed 101 patients with clinical stage II-III rectal cancer who underwent neoadjuvant therapy and found, consistent with global literature, that approximately 20% of patients achieved a complete response, while tumor size was inversely correlated with the probability of achieving a complete clinical response.

INTRODUCTION

Globally, colorectal cancer (CRC) remains a leading cause of cancer-related morbidity and mortality. GLOBOCAN estimates in 2022 reported over 1.9 million new cases and approximately 935000 deaths annually, ranking CRC as the third most commonly diagnosed malignancy and the second leading cause of cancer-related death worldwide. Rectal cancer, which constitutes one third of the total incidence of CRC accounted of 729702 cases and 538138 deaths worldwide (male and female cases combined)[1]. Although incidence is highest among individuals aged 50 years or older, contemporary population-based registries have demonstrated a sustained increase in younger patients, particularly those aged 40-49 years[2]. In Mexico, the disease burden is characterized by a high proportion of diagnoses at advanced stages. Recent international studies indicate that a substantial proportion of patients with rectal cancer continue to present with locally advanced disease, highlighting the ongoing clinical challenges that necessitate multimodal treatment strategies[3-5]. In these cohorts, most patients were older than 50 years, with a slight male predominance, and presented with stage III or IV disease at diagnosis, underscoring the need for effective strategies to achieve locoregional control[2,6-9].

Locally advanced rectal cancer (LARC) represents a distinct clinical challenge within the spectrum of colorectal malignancies, owing to its unique anatomical characteristics, locoregional patterns of spread, and elevated risk of recurrence[4,10,11]. Anatomically, the rectum comprises the distal 16 cm of the large bowel and is conventionally divided into upper, middle, and lower segments. Variations in mesorectal anatomy and lymphatic drainage directly influence tumor dissemination and oncologic outcomes[12-15]. From a histopathological standpoint, more than 95% of rectal tumors are adenocarcinomas. LARC—defined by transmural invasion of the primary tumor (T3-T4), regional lymph node involvement, and/or threatened mesorectal fascia—is associated with a substantially increased risk of local recurrence and systemic dissemination when managed with surgery alone[16,17]. Unlike colon cancer, the rectum lacks a serosal covering and maintains close anatomical relationships with adjacent pelvic organs, limiting the ability to achieve wide surgical margins and providing a strong rationale for a multimodal therapeutic approach[18].

Neoadjuvant treatment—consisting of radiotherapy with or without chemotherapy administered prior to surgical resection—has become a cornerstone in the management of LARC[19-21]. Its primary objectives include reducing locoregional recurrence, improving tumor resectability and in selected cases, enabling organ-preservation strategies[7]. Randomized clinical trials and high-impact studies published over the past five years have demonstrated that intensification and re-sequencing of systemic therapy within the framework of total neoadjuvant therapy (TNT) increase complete response rates without compromising oncologic outcomes[11,13,22,23]. Tumor response is typically assessed between 6 and 12 weeks after completion of neoadjuvant treatment, a time window that maximizes tumor regression. Pathological complete response (pCR) is defined as the absence of viable tumor cells in the surgical specimen (ypT0N0), whereas complete clinical response (cCR) refers to the absence of detectable disease on digital rectal examination, endoscopy, and pelvic magnetic resonance imaging (MRI)[16]. Reported pCR rates range from 10% to 30%, depending on the treatment regimen employed, while cCR has gained increasing clinical relevance by enabling non-operative management strategies. In this setting, the watch-and-wait approach has emerged as a management strategy based on close surveillance of patients with sustained cCR, aiming to avoid radical surgery and preserve anorectal function. Prospective studies and international databases have demonstrated that a substantial proportion of patients can safely retard definitive surgical resection without compromising overall survival, provided that careful patient selection and structured follow-up protocols are implemented[24,25]. Nevertheless, early local recurrence remains a central concern, with most events occurring within the first two years of follow-up.

Despite these advances, considerable heterogeneity persists in the therapeutic regimens and sequencing strategies employed within TNT, making it challenging to identify the optimal approach to maximize the likelihood of achieving cCR. This issue is particularly relevant in Latin American populations, where delayed diagnosis and higher tumor burden are frequently observed, as reported in recent Mexican literature[2,6]. Identifying factors associated with response and evaluating response durability have direct implications for multidisciplinary decision-making, patient quality of life, and the potential avoidance of high-morbidity surgical procedures.

Against this background, the present study analyzes an institutional cohort of Mexican patients with LARC treated with neoadjuvant therapy to describe the rate of cCR, its characteristics, local recurrence patterns, and factors associated with achieving cCR.

MATERIALS AND METHODS

Ethical statements

Patients diagnosed at the Oncology Hospital of the National Medical Center between 2022 and 2024 were included. The Mexican Institute for Social Security Ethics Committee reviewed and approved the study (Approval No. R-2025-3602-022).

Study design and population

A retrospective observational cohort study was conducted, including consecutive patients diagnosed with LARC, defined as clinical stage II or III, who were treated with neoadjuvant therapy at a tertiary referral center. The study period encompassed all patients who completed neoadjuvant treatment and had a documented post-treatment response assessment available for analysis (Figure 1).

Figure 1 Flow diagram of the cohort and assessment of complete response. The diagram illustrates patient selection, neoadjuvant treatment, post-treatment response assessment, and clinical outcomes. It shows the proportion of patients achieving a complete response, including those with complete clinical response managed with a watch-and-wait strategy and those with pathological complete response confirmed after surgery, as well as the occurrence of local recurrence during follow-up. LARC: Locally advanced rectal cancer; MRI: Magnetic resonance imaging; cCR: Complete clinical response; pCR: Pathological complete response.

LARC diagnosis

All recruited patients had a complete clinical history confirmed by a colonoscopy and rectal biopsy for histological evaluation, MRI, digital rectal examination and computed tomography. In blood, coagulation tests, leukocyte counts, liver and renal function tests, serum carcinoembryonic antigen and serum electrolytes were performed.

Radiotherapy

Short-course radiotherapy was given as 25 Gray in five fractions (5 Gray per session) over 5-8 days, planning target volume by three-dimensional conformal radiotherapy technique. Long-course radiotherapy was given as 50.4 Gray in 28 fractions at 1.8 Gray per session over 5-6 weeks. Both treatments were given on an Elekta Infinity.

Neoadjuvant chemotherapy

After completion of radiotherapy, most patients received neoadjuvant chemotherapy according to institutional protocols, most commonly CAPOX.

Neoadjuvant treatment and response assessment

Following completion of therapy, tumor response was systematically evaluated using a combination of clinical examination, endoscopic assessment and imaging studies. The primary outcome was cCR, defined as the absence of detectable residual tumor on clinical, endoscopic, and radiological evaluation, followed by management with an active surveillance (watch-and-wait) strategy. The pCR was defined as the absence of viable tumor cells confirmed histologically in the surgical specimen following tumor resection. An overall complete response rate (cCR or pCR) is reported descriptively to contextualize response frequency; inferential analyses were focused on cCR.

Variables and data collection

Demographic, clinical and tumor-related variables were collected for the entire cohort, including age, sex, clinical stage, tumor size, tumor location, distance from the anal verge, and baseline biochemical parameters. In the subgroup of patients achieving cCR and managed with watch-and-wait, additional variables were analyzed, including functional outcomes, neoadjuvant treatment characteristics, and the occurrence of local recurrence during follow-up. For patients who underwent surgery, pCR status was recorded and reported separately.

Follow-up and definition of local recurrence

Patients managed with a watch-and-wait strategy after cCR were followed at regular intervals in accordance with institutional surveillance protocols, which included monitoring every 3-4 months for up to 3 years after treatment. Local recurrence was defined as the reappearance of tumor within the rectum or pelvis, confirmed by clinical evaluation, endoscopic findings or radiological imaging during follow-up.

Statistical analysis

Continuous variables were summarized as mean ± SD or median (interquartile range), as appropriate, based on data distribution. Categorical variables were reported as absolute n (%). An exploratory univariate analysis was performed to identify factors associated with cCR, using appropriate statistical tests according to variable type. Response durability and local recurrence-free survival among patients managed with watch-and-wait after cCR were analyzed using the Kaplan-Meier method. A two-sided P value < 0.05 was considered statistically significant. All analyses were performed using standard statistical software.

RESULTS

Clinical and treatment characteristics of the study cohort

The study cohort comprised 101 patients with LARC, with a mean age of 61.7 ± 13.3 years (range: 23-85 years). There were 55 men (54.5%) and 46 women (45.5%). Most patients presented with clinical stage III disease (81.2%), while 18.8% were diagnosed at stage II. Mean tumor size was 7.28 ± 2.52 cm, and tumors ≥ 5 cm accounted for 78.2% of cases. Tumors were most frequently located in the lower rectum (39.6%), followed by the middle (38.6%) and upper rectum (21.8%). The mean distance from the anal verge was 6.91 ± 3.74 cm. All patients received neoadjuvant radiotherapy, and 82 (81.2%) received concomitant chemotherapy (Table 1).

Table 1 Clinical, tumor, and neoadjuvant treatment characteristics of the overall cohort (n = 101).

Characteristic	Value
Age (years), mean ± SD (range)	61.7 ± 13.3 (23-85)
Sex
Male	55 (54.5)
Female	46 (45.5)
Clinical stage
II	19 (18.8)
III	82 (81.2)
Tumor size (cm)	7.28 ± 2.52
< 5	22 (21.8)
≥ 5	79 (78.2)
Tumor location
Lower rectum	40 (39.6)
Middle rectum	39 (38.6)
Upper rectum	22 (21.8)
Distance from anal verge (cm)	6.91 ± 3.74
Hemoglobin < 10 g/dL	34 (33.7)
CEA ≥ 5 ng/mL	50 (49.5)
Albumin (g/dL)	4.0 ± 0.59
Leukocytes (× 103/μL)	7.28 ± 5.03
Neoadjuvant radiotherapy	101 (100)
Neoadjuvant chemotherapy
Yes	82 (81.2)
No	19 (18.8)
Complete response	20 (19.8)

Data were showed as mean ± SD or n (%). Complete response is reported descriptively and defined as complete clinical response managed with watch-and-wait or pathological complete response confirmed after surgery. Inferential analyses in this study were focused on complete clinical response. CEA: Carcinoembryonic antigen.

Complete response rate and follow-up outcomes

An overall cCR managed with watch-and-wait or pCR confirmed after surgery was observed in 19.8% of the cohort (n = 20). Among complete responders, 15 patients (75.0%) achieved cCR and were managed with an active surveillance (watch-and-wait) strategy, whereas 5 patients (25.0%) had pCR confirmed following resection.

During follow-up of complete responders (n = 20), two local recurrences were documented (10.0%). Both events occurred within the first 16 months after response assessment (11.0 months and 15.9 months) and were observed exclusively among patients managed with watch-and-wait after cCR (2/15; 13.3%; Table 2). The remaining 18 complete responders remained free of local recurrence at last follow-up. The Kaplan-Meier curve for local recurrence among cCR watch-and-wait patients is shown in Figure 2, demonstrating that recurrence events clustered early during follow-up; median local recurrence-free survival was not reached due to the low number of events.

Figure 2 Kaplan-Meier curve for local recurrence. The graph shows the cumulative probability of local recurrence-free survival among complete responders (n = 20). Two local recurrence events were registered within the follow-up period (11 months and 15.9 months). The median recurrence-free survival was not reached due to the small number of events.

Table 2 Patients achieving complete response after neoadjuvant (n = 20).

Variable	Value
Age (years)	61.6 ± 12.9
Sex
Male	12 (60.0)
Female	8 (40.0)
ECOG performance status
0	11 (55.0)
1	8 (40.0)
≥ 2	1 (5.0)
Distance from anal verge (cm)	5.0 (1.5-6.3)
Tumor length (cm)	5.8 ± 2.4
Baseline CEA (ng/mL)	3.2 (1.8-6.4)
Post-neoadjuvant CEA (ng/mL)	1.9 ± 0.8
Clinical T stage
≥ T3	17 (85.0)
≤ T2	3 (15.0)
Clinical N stage
Positive	14 (70.0)
Negative	6 (30.0)
Radiotherapy type
Long-course radiotherapy	18 (90.0)
Short-course radiotherapy	2 (10.0)
Neoadjuvant chemotherapy regimen
CAPOX	14 (70.0)
Capecitabine	3 (15.0)
FOLFOX	3 (15.0)
cCR	15 (75.0)
pCR	5 (25.0)
Local recurrence during follow-up among complete responders	(n = 20)
Yes	2 (10.0)
No	18 (90.0)

Data were showed as mean ± SD, median (interquartile range), or n (%). Both local recurrence events occurred exclusively in the complete clinical response watch-and-wait subgroup (2/15; 13.3%). ECOG: Eastern Cooperative Oncology Group; CEA: Carcinoembryonic antigen; cCR: Complete clinical response; pCR: Pathological complete response.

Exploratory analysis of factors associated with cCR

An exploratory univariate analysis was performed to evaluate factors associated with cCR in the overall cohort (n = 101). Tumor size was the only variable showing a statistically significant association with cCR, with a decreasing likelihood of cCR as tumor size increased (OR = 0.77 per centimeter; 95%CI: 0.61-0.96; P = 0.028). Other variables—including sex, clinical stage, tumor location, biochemical parameters, and neoadjuvant treatment category (chemoradiotherapy vs radiotherapy alone)—were not significantly associated with cCR in univariate models (Table 3). Continuous variables compared using Student’s t test showed no significant differences for age at diagnosis, distance from the anal verge, or degree of luminal stenosis. Although sex showed a statistically significant association with cCR in the χ² test (P = 0.014), this finding was not supported by the odds ratio-based analysis. These univariate results are intended for signal detection and hypothesis generation and should not be interpreted as demonstrating independent predictive effects.

Table 3 Exploratory univariate analysis of factors associated with complete clinical response after neoadjuvant treatment (n = 101).

Variable	Measure of association (OR)	95%CI	P value
Categorical variables (univariate logistic regression)
Sex (male vs female)	1.19	0.35-4.05	0.77
Hemoglobin (< 10 g/dL vs ≥ 10 g/dL)	1.14	0.30-4.32	0.97
CEA (> 5 ng/mL vs ≤ 5 ng/mL)	0.98	0.76-1.17	0.83
Albumin (≤ 3 g/dL vs > 3 g/dL)	0.59	0.20-1.75	0.67
Tumor size (per cm increase)	0.77	0.61-0.96	0.028
Tumor location (lower rectum vs others)	0.98	0.20-4.33	0.89
Clinical stage (III vs II)	1.32	0.59-2.97	0.49
Neoadjuvant treatment (CRT vs RT alone)	0.70	0.12-4.04	0.86
Continuous variables (Student’s t test)
Age at diagnosis (years)	0.0111	-20.6 to 20.8	0.991
Distance from anal verge (cm)	1.6551	-4.8 to 5.3	0.215
Degree of luminal stenosis (%)	0.8201	-31.4 to 249.7	0.382

¹t statistic.

All analyses are exploratory and based on univariate models. Given the limited number of complete clinical response events (n = 15), multivariable modeling was not performed to avoid overfitting. Therefore, results should not be interpreted as demonstrating independent predictive effects. CEA: Carcinoembryonic antigen; CRT: Chemoradiotherapy; RT: Radiotherapy.

DISCUSSION

There was a significant overall complete response rate of 19.8% among patients with LARC treated with neoadjuvant therapy (cCR managed with watch-and-wait or pCR confirmed following surgery). Most complete responders showed cCR and were treated with an active monitoring strategy, and fewer reported pCR confirmed post-resection. These response rates, in general, are within the range reported in the current series of randomized trials, in which the frequency of complete response remains dependent on treatment regimen, sequencing, and response assessment criteria[11,22].

Another critical aspect of this study concerns response duration and the trend in local recurrence. In modern populations managed with watch-and-wait strategies following chemoradiotherapy or TNT, most local regrowth events occur within the first two years of follow-up, with a median time to regrowth of 9 months (International Watch & Wait Database), and conditional recurrence-free survival is markedly improved thereafter[7,16,23,24,26].

A recent meta-analysis has reported cumulative local recurrence rates of 15%-20% at mid-term follow-up[25]. In keeping with these data, we recorded two local recurrences of the same duration in our cohort of complete responders (2/20; 10.0%), both occurring within the first 16 months after response evaluation and importantly, exclusively within the cCR watch-and-wait subgroup (2/15; 13.3%). However, this premature clustering, despite low incidence and failure to achieve median local recurrence-free survival, was consistent with previous studies and confirms the clinical significance of close follow-up in the initial two years in centers that have been operationalizing the watch-and-wait protocol[23,24,26,27].

In this cohort, tumor size was the only variable associated with cCR in an exploratory univariate model, in which larger tumors were associated with a lower likelihood of cCR (OR = 0.77 per centimeter). This association is clinically plausible because smaller tumors are likely to produce a lower tumor burden and a more uniform response to treatment. Nevertheless, this finding is not independent of other outcomes due to the univariate design and the low incidence of cCR events; cautious inference should therefore be made. Instead, it suggests that tumor size may be a pragmatic, easily measurable surrogate for a multidisciplinary dialogue on the organ-preservation mechanisms in selected patients. In univariate models, however, other evaluated variables such as sex, clinical stage, tumor location, biochemical parameters and treatment type (chemoradiotherapy vs radiotherapy alone) were not consistently associated with cCR. The discordance between the χ² test and the odds ratio-based analysis for sex highlights the power of spurious signals in exploratory subgroup comparisons and the importance of avoiding causal inferences from unadjusted analyses, particularly when event counts are low.

This study suffers from limitations inherent to a retrospective study design, particularly the relatively small number of responders (n = 20, 15 cCR events). This small subgroup of subjects significantly reduced the statistical power of our analysis, preventing robust multivariable modeling and limiting our ability to assess interactions between clinical and tumor-related covariates. Thus, any associations identified should be treated with caution as explorative findings. Furthermore, systematic molecular characterization was not available, which should have fine-tuned patient selection and increased interpretability. Still, the strengths of this work include analysis of a real-world cohort with distinct response criteria, an explicit distinction between cCR and pCR, and standardized follow-up in watch-and-wait patients, which lend credence to its applicability to real-world clinical practice. Collectively, these findings reinforce the potential of neoadjuvant therapy to achieve clinically meaningful complete responses in a narrow group of patients with LARC and complement real-world data from a Mexican population. The detection of the influence of tumor size on cCR, along with the low local recurrence rate noted with close surveillance, highlights the utility of including basic clinical variables in decision-making among all areas of MDT. Further prospective multicenter studies including both radiological and molecular biomarkers will also be critical to optimizing patient selection in this context and further solidifying organ-preservation strategies.

CONCLUSION

In our cohort of patients with LARC treated with neoadjuvant therapy, an overall complete response (with cCR managed with watch-and-wait or pCR confirmed after surgery) occurred in 1 in 5 patients; most complete responders achieved cCR and were enrolled in structured surveillance. Durability of response was favorable, with a low rate of early local recurrence, indicating the viability and safety of close monitoring in selected individuals treated in organized monitoring systems. The only clinical covariate strongly associated with the likelihood of a complete response in exploratory univariate analysis was tumor size; hence, the use of simple and reproducible clinical parameters might guide the selection of candidates for organ preservation efforts. Despite limitations, such as the retrospective design and the small size of the complete response subgroup (n = 20), the findings provide clinically relevant evidence in a real-world setting and support the role of neoadjuvant therapy and observational strategies when implemented in experienced centers. Future large-scale, multicenter prospective studies that incorporate radiologic and molecular factors into selection criteria will be needed to identify patients most likely to achieve and maintain a cCR.