Quality of life following total neoadjuvant therapy for rectal cancer

Abstract

Purpose

This study aimed to assess the health-related quality of life (HRQoL) in patients with locally advanced rectal cancer (LARC) undergoing total neoadjuvant therapy (TNT), comparing outcomes with the German general population and colorectal cancer (CRC) patients treated with curative intent.

Methods

In a multicenter, cross-sectional study within the “TNTox” study framework (DRKS 00033000), EORTC QLQ-C30 and QLQ-CR29 questionnaires were distributed to LARC patients who had completed TNT. Mean reference values were compared descriptively, and further exploratory comparisons based on clinical features were performed.

Results

The study included responses from 72 patients. Compared to the German general population, a reduction in mean HRQoL across most domains was observed; the strongest effect was observed for role functioning (− 28.7 points, Cohen’s d = − 0.95), social functioning (− 25.3 points, d = − 0.89), and for diarrhea (+ 9.9 points, d = 0.80). General HRQoL was similar to that of CRC patients following curative treatment. However, some symptom scores, notably fecal incontinence (+ 13.4 points, d = 0.52), impotence (+ 29.0 points, d = 0.73), and dyspareunia (+ 10.4 points, d = 0.40) appeared to be higher. Significant factors associated with HRQoL included the presence of chronic treatment-related toxicity and duration of TNT; no major differences were observed between patients with or without NOM or stoma.

Conclusion

LARC patients undergoing TNT showed comparable HRQoL outcomes to CRC patients treated with curative intent, but with reductions when compared to the general population. The presence of chronic toxicity significantly impacts HRQoL. Survivors may experience HRQoL impairments post-TNT, underscoring the necessity for ongoing management of chronic toxicity tailored to their needs.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00432-025-06347-y.

Introduction

The prognosis of locally advanced rectal cancer (LARC) has substantially improved by modern resection techniques and emerging multimodal treatment approaches within the last decades (Glynne-Jones et al. 2017; Keller et al. 2020). Within these patients, total neoadjuvant therapy (TNT) is currently widely applied in the presence of certain risk factors, improving both local control and reducing metastatic spread (Bahadoer et al. 2021; Conroy et al. 2021; Fokas et al. 2019; Garcia-Aguilar et al. 2022). This treatment intensification adds preoperative chemotherapy to neoadjuvant radiotherapy (RT) and further increases complete response (CR) rates to 25–65% compared to 10–15% after standard neoadjuvant chemoradiotherapy (CRT) (Kasi et al. 2020; Rödel et al. 2015; Wurschi et al. 2023), enabling non-operative management (NOM) in an increasing number of patients with complete response. Radical resection is known to be associated with reduced health-related quality of life (HRQoL) (Burch et al. 2021). NOM might be associated with better QoL through preserved (anorectal) organ function (Gilbert et al. 2022; Jones et al. 2020). Similarly, a diverting ostomy might deteriorate QoL (Lawday et al. 2021). However, the toxicity profile after TNT varies due to more intense chemotherapy treatment, probably affecting QoL. Chemotherapy within TNT is usually based on 5-fluorouracil and oxaliplatin containing regimens, such as FOLFOX or CAPOX (Ochiai et al. 2024). Oxaliplatin may cause persistent peripheral neurological adverse effects, such as polyneuropathy (PNP) or autonomic dysregulation, relevantly impairing instrumental activities of daily living and QoL (Jordan et al. 2020; Loprinzi et al. 2020).

Given the relatively good prognosis of LARC patients after TNT with 5-year overall survival rates > 74% (Bahadoer et al. 2021; Garcia-Aguilar et al. 2022; Petrelli et al. 2020; Fokas et al. 2022), a relevant proportion of patients might face long-term toxicity that could potentially impact their QoL. It is crucial to identify potential factors that may impair HRQoL in these patients. To date, only limited data on HRQoL are available for this new therapy (Baird et al. 2023). NOM omits radical resection, thus requiring more frequent follow-up visits. It could, on the other hand, also result in increased (psychological) stress due to uncertainty and anxiety regarding tumor control (Liu et al. 2021). For patients following radical resection of colorectal cancers, an improved HRQoL, known as ‘post-traumatic growth’, is described (Kim et al. 2021; Scherer-Trame et al. 2022). Whether this has an inverse effect on HRQoL in the NOM cohort has not been reported so far.

We thus aim to analyze a cross-sectional cohort of LARC patients after TNT within the so-called “TNTox” study (Wurschi et al. 2024) and compare their quality of life with reference data from the German general population and colorectal cancer patients treated with curative intent. In addition, we sought to assess clinically relevant factors potentially associated with QoL.

Material and methods

We conducted a multicenter analysis within the “Young DEGRO” working group of the German Society for Radiation Oncology (DEGRO) at 23 hospitals in Germany and Austria (Wurschi et al. 2024). During follow-up, the European Organisation for Research and Treatment of Cancer Quality of Life questionnaires QLQ-C30 (EORTC QLQ-C30) (Aaronson et al. 1993) and EORTC QLQ-CR29 (Whistance et al. 2009; Gujral et al. 2007) were distributed prospectively to all eligible patients at 10/23 participating centers, using a cross-sectional approach. Written informed consent was obtained from all participants. The study was approved by the local ethics committee of the Faculty of Medicine, Jena University Hospital (2023-3042-Bef), and by each participating center's ethics committee and adhered to the Declaration of Helsinki. The trial protocol was prospectively registered with the German Clinical Trials Registry (DRKS, No. 00033000) and accredited by the Radiation Oncology Working Group of the German Cancer Society (Arbeitsgemeinschaft Radiologische Onkologie, ARO). All analyses are conducted according to the STROBE criteria (Elm et al. 2007).

Eligible patients were diagnosed with localized rectal cancer (T2-4 N0-2 M0) and underwent neoadjuvant RT, consisting of short-course radiotherapy (SCRT) or long-course chemoradiotherapy (LCRT), followed by consolidation chemotherapy with curative intent between 2015 and 2024. Clinical data were collected retrospectively. Follow-up data were gathered through routine oncological follow-up visits according to institutional standards. Data collection and management were performed using REDCap electronic data capture tools (Harris et al. 2019, 2009) hosted at Jena University Hospital, Jena, Germany.

Questionnaires

The EORTC QLQ-C30 includes 30 questions that are assigned to five functioning scales (physical, role, cognitive, emotional, and social functioning), three symptom scales (fatigue, pain, and nausea and vomiting), a global health status scale and additional single items assessing commonly reported symptoms (dyspnea, loss of appetite, insomnia, constipation and diarrhea, financial impact of the disease). The QLQ-CR29 module comprises 29 additional specific questions for colorectal cancer patients on 4 multi-item scales and 18 single-items (Whistance et al. 2009), which were subsumed to 4 functioning scales (anxiety, weight loss, body image, sexual interest) and 17 symptom scales (urinary frequency, urinary incontinence, dysuria, abdominal pain, buttock pain, bloating, blood/mucus in stool, dry mouth, hair loss, taste alteration, impotence/dyspareunia, flatulence, fecal incontinence, stool frequency, sore skin, embarrassment). These subscales (range, 0–100) were calculated for both questionnaires (from the corresponding single item values) according to the EORTC manuals (Fayers et al. 2001). Higher scores on functioning scales represent better function, whereas higher symptom scores indicate a high level of symptoms (Aaronson et al. 1993). QLQ-C30 subscales were summarized in the established “Summary Score” (Giesinger et al. 2016). For the QLQ-CR29, no symptom-related summarization of single items is officially validated (Hout et al. 2019). To ease comparison, the items 49–54 were summarized to the “Defecation/Stoma-related problems” scale, which was introduced and internally validated by Stiggelbout et al. (2016).

Endpoints and definitions

The above-defined QLQ-C30 and QLQ-CR29 subscales served as co-primary endpoints. Given the large number of single-item scales, explorative testing was limited to the most relevant multi-item scales reflecting general QoL, i.e., the “Global Health Scale”, the “Summary Score”, as well as bowel-related symptoms (i.e., “Defecation/Stoma-related problems” and “Blood/mucus in stool”) together with the related “Body image” scale.

Within the prospective study protocol, clinical features, including demographic factors, such as age, sex, presence of chronic treatment-related toxicity, the Karnofsky performance status (KPS), the duration of TNT in months, recurrent disease, as well as NOM and colostomy, were prespecified as factors for further explorative comparison. Chronic toxicity was assessed according to CTCAE v5.0 and was considered symptomatic if graded ≥ 2. For these analyses, all patients undergoing a “watch and wait” approach after CR or local excision were considered as “NOM” patients.

Statistical analysis

All analyses are exploratory. HRQoL values were primarily reported descriptively with mean (± standard deviation, SD) and median (first and third quartiles). Mean scores of the subscales were compared against reference values for the German general population (Hinz et al. 2014) and EORTC reference values for colorectal cancer (CRC) patients after treatment with curative intent (Whistance et al. 2009). To compare the patients’ scores with those of the general population, we used normative data taken from a large representative general population study for the QLQ-C30 (Hinz et al. 2014). The following strategy was used to mitigate demographic differences regarding sex and age distribution: In the publication of the normative scores, we used the women’s and men’s mean scores and standard deviations of the age decades as reported in the normative table. The age decades were weighted according to the frequencies of the age decades in the cancer sample. For example, the percentage of the four male patients in the age group 40–49 years was 5.6% in this study. These percentages were taken as the weighting factors for calculating the general population mean value of the QLQ-C30 scores.

A difference in mean values of ≥ 10 points was considered clinically relevant (Osoba et al. 1998; Musoro et al. 2020). The effect size for these comparisons was estimated with Cohen’s d (New and Jersey 1988). Cohen’s d > 0.2 was considered as weak, d > 0.5 as moderate, and d > 0.8 as a strong effect, respectively (New and Jersey 1988; Sullivan and Feinn 2012).

Spearman’s correlation was calculated for the evaluation of continuous variables, whereas known group comparisons were performed with Mann–Whitney U tests. Following standard practice (Hout et al. 2019), these comparisons were limited to the Global Health Status (GHS) and the QLQ-C30 summary score (Giesinger et al. 2016) of the QLQ-C30 questionnaire. For the QLQ-CR29, the multi-item scales “Body Image” and “Blood/Mucus in stool” were used, along with “Defecation/Stoma-related problems” (items 49–54). Internal consistency of the latter multi-item scale was assessed using Cronbach’s α and McDonald’s ω prior to the application.

All 95% confidence intervals (95% CI) were obtained via nonparametric bootstrapping with 1000 resamples using the percentile method. Given the exploratory nature of this study, primary analyses were performed using unadjusted two-sided p-values with a significance threshold of p < 0.05. To address the issue of multiple testing, test families were defined a priori, and Holm’s sequential Bonferroni procedure was applied within each known-group comparison across the respective 5 HRQoL scales as a sensitivity analysis (adjusted threshold p < 0.01). All analyses were conducted with SPSS 29.0 (IBM SPSS Statistics, Armonk, NY / USA) as well as JASP v0.19.3 (JASP Team 2025). Excel v16.96 (Microsoft Inc., Redmond, WA / USA) was used for further visualization.

Results

Patient’s characteristics and treatments

For this analysis, questionnaires from 72/295 patients (24.4%) across 10 centers participating (Supplement S1) in the TNTox study were available. Their median age at diagnosis was 65 years (Q1-Q3: (56.5–69.0)), and among them, 54 (75.0%) were men. LCRT was applied in the majority of the patients (66/72, 91.7%). A median of 6.0 (Q1-Q3: 3.0–9.0) cycles of consolidation chemotherapy (standardized to FOLFOX-cycles, i.e., q2w) was administered with a median TNT-duration of 4.4 (Q1-Q3: 2.9–5.4) months. The questionnaires were obtained after a median interval of 17.0 (Q1-Q3: 9.3–34.0) months after completion of TNT. At that time, 26 out of 71 patients (16.9%) presented with a diverting ostomy, and 20 out of 72 patients (27.8%) were managed non-operatively. A detailed description of the characteristics is provided in Tables 1A and 1B. Additional information regarding treatment failures with related treatments and chronic toxicity at follow-up can be found in the Supplements S2 and S3. Overall, symptomatic chronic toxicity (CTCAE grade ≥ 2) was reported for 34 out of 69 (49.3%) patients. Most frequent CTCAE grade ≥ 2 events comprised PNP in 19 out of 69 (27.5%) patients and chronic fatigue in 9 out of 68 (13.2%) patients. Furthermore, CTCAE grade ≥ 2 erectile dysfunction was reported in 14 out of 45 (31.1%) men and dyspareunia in one out of 15 (6.7%) women.

Table 1.

Characteristics of included patients. Categorical variables (A) and continuous variables (B) with mean and standard deviation (SD) and median, 1st and 3rd quartile, are shown separately. Due to rounding, percentages may not add up to exactly 100%. Note that confidence intervals are based on 1000 bootstrap replicates

Variable	N	n (%)
(A) Categorical variables
Sex	72	Male: 54 (75.0%)		Female: 18 (25.0%)
Chemoradiotherapy (LCRT)	72	Yes: 66 (91.7%)		No: 6 (8.3%)
Non-operative management (NOM)	72	Yes: 20 (27.8%)		No: 52 (72.2%)
Diverting ostomy at follow-up (“stoma”)	71	Yes: 26 (36.6%)		No: 45 (63.4%)
Relapse	71	Yes: 12 (16.9%)		No: 59 (83.1%)
ESMO Risk Classification	72	Early: 1 (1.4%)	Intermediate: 22 (30.6%)	Bad: 15 (20.8%)	Advanced: 34 (47.2%)
TNM: T Stage	72	T1: 2 (2.8%)	T2: 3 (4.2%)	T3a/b: 36 (50.0%)	T3c/d: 21 (29.2%)	T4: 10 (13.9%)
TNM: N Stage	72	N0: 8 (11.1%)	N1: 26 (36.1%)	N2: 27 (37.5%)	N + (not specified): 11 (15.3%)
TNM: M Stage	72	M0: 72 (100%)
Localization from the anal verge	72	0–6 cm 38 (52.8%)	6–12 cm 32 (44.4%)		> 12 cm 2 (2.8%)
Infiltration of the mesorectal fascia (MRF +)	70	< 1 mm / MRF +  33 (47.1%)	1–2 mm / MRF threatened 9 (12.9%)		> 2 mm / MRF clear 28 (40.0)
Extramural vascular invasion (EMVI)	51	Yes: 14 (27.5%)		No: 37 (72.5%)
Involved lateral pelvic nodes	64	Yes: 13 (20.3%)		No: 37 (79.7%)
Any chronic toxicity at follow-up (CTCAE grade)*	69	≥ 2		34 (49.3%)
		≥ 3		18 (26.1%)

	n	Mean	95% Confidence Interval Mean		SD	Median	Q1	Q3
			Lower	Upper
(B) Continuous Variables (N = 72)
Age (years)	72	63.3	61.0	65.4	9.5	65.0	57.5	69.0
TNT duration (months)	72	4.4	4.1	4.8	1.5	4.4	2.9	5.4
Follow-up (months since end of TNT)	72	22.4	18.3	26.8	18.0	17.0	9.8	34.0
Chemotherapy cycles (standardized to FOLFOX q2w)	72	6.0	5.5	6.7	2.7	6.0	3.0	9.0
Karnofsky Performance Status (KPS): Baseline	72	91.4	89.7	92.9	7.0	90.0	90.0	100.0
Karnofsky Performance Status (KPS): Follow-up	65	86.0	84.0	88.2	8.6	90.0	80.0	90.0

*A detailed description of the distinct toxicity grades is provided in Supplement 3

Quality of life scores

HRQoL scores are provided for every subscale of the QLQ-C30 and QLQ-CR29 questionnaires in Table 2. A detailed description of the QLQ-CR29 single items for patients with and without a stoma is provided in Supplement S4. The QLQ-CR29 multi-item scale “Defecation/Stoma related problems” was calculated according to Stiggelbout et al. (2016). We found a good internal consistency with Cronbach’s α of 0.847 (95% CI: 0.771–0.901) and McDonald’s ω of 0.845 (95% CI: 0.758–0.899) for this scale in our cohort (Supplement S5). Moderate correlations with most of the established QLQ-C30 and QLQ-CR29 scales related to bowel function/defecation symptoms were observed (Spearman’s ρ 0.45–0.59). Only the correlation with the “Blood and Mucus in stool” scale was not statistically significant (p = 0.068). For details, see Supplement S5. Given its good internal consistency and clinically plausible correlations, this new multi-item scale was deemed appropriate for our cohort and was subsequently used for further analyses.

Table 2.

Summarized QLQ-C30 scores (A) and QLQ-CR29 scores (B) of functioning scores and symptom scores. Descriptive parameters, i.e., the number of questionnaires (n), mean together with standard deviation (SD), and 95% confidence interval (CI), are provided (B). 95% CIs are based on 1000 bootstrap samples. Higher scores in function scales indicate better functioning, whereas higher scores in symptom scales indicate more symptoms

		Mean	95% Confidence Interval Mean		SD	Median	Q1	Q3
	n		Lower	Upper
(A) QLQ-C30 Scores
Functioning scales
Physical Functioning	72	76.7	70.9	81.6	23.6	80.0	60.0	100.0
Role Functioning	70	61.7	54.3	68.6	31.6	66.7	33.3	95.8
Emotional Functioning	70	75.4	68.3	81.3	26.7	83.3	58.3	100.0
Cognitive Functioning	69	82.1	76.3	87.2	23.3	100.0	66.7	100.0
Social Functioning	68	68.1	60.8	75.0	31.2	66.7	50.0	100.0
Global Health	69	64.1	58.6	69.2	23.1	66.7	50.0	83.3
Summary Score	64	77.8	73.0	82.3	19.0	81.8	64.5	92.5
Symptom scales
Fatigue	71	29.0	23.2	35.1	26.4	22.2	5.6	44.4
Nausea/Vomiting	70	5.0	2.6	8.3	12.2	0.0	0.0	0.0
Pain	70	19.8	14.3	26.0	25.9	8.3	0.0	33.3
Dyspnea	71	16.9	12.2	22.5	23.8	0.0	0.0	33.3
Insomnia	71	28.6	21.1	36.6	33.5	33.3	0.0	66.7
Appetite Loss	71	13.1	6.6	19.3	27.9	0.0	0.0	0.0
Constipation	71	13.1	7.5	19.2	24.9	0.0	0.0	33.3
Diarrhea	69	21.3	14.0	29.0	30.8	0.0	0.0	33.3
Financial Difficulties	68	17.6	11.3	24.5	29.1	0.0	0.0	33.3
(B) QLQ-CR29 Scores
Functioning scales
Body Image	69	77.0	71.3	82.9	25.6	88.9	66.7	100.0
Weight Loss	69	78.3	72.0	84.1	25.5	100.0	66.7	100.0
Anxiety	69	60.4	52.7	68.6	33.0	66.7	33.3	100.0
Sexual Interest (Men)	48	42.4	34.0	50.7	32.1	33.3	25.0	66.7
Sexual Interest (Women)	15	22.2	11.1	35.6	24.1	33.3	0.0	33.3
Defecation/Stoma-related problems	64	29.7	24.6	35.5	23.1	27.8	11.1	44.4
Symptom scales
Urinary Frequency	69	37.9	31.9	43.7	25.9	33.3	16.7	50.0
Urinary Incontinence	67	11.9	7.5	15.9	19.0	0.0	0.0	33.3
Dysuria	71	7.5	3.8	12.2	18.9	0.0	0.0	0.0
Abdominal Pain	71	11.7	7.0	16.4	20.4	0.0	0.0	33.3
Buttock Pain	71	23.0	17.4	28.6	26.8	0.0	0.0	33.3
Bloating	70	21.4	15.2	28.1	26.6	0.0	0.0	33.3
Blood/Mucus in Stool	65	7.9	5.1	11.0	12.9	0.0	0.0	16.7
Dry Mouth	71	16.9	11.3	22.5	24.5	0.0	0.0	33.3
Hair Loss	71	16.0	10.3	22.1	25.7	0.0	0.0	33.3
Taste Alteration	71	15.5	8.9	23.0	29.2	0.0	0.0	33.3
Impotence (Men)	46	63.0	51.4	73.9	38.6	66.7	33.3	100.0
Dyspareunia (Women)	12	19.4	5.6	36.1	30.0	0.0	0.0	33.3
Stoma Care Problems	25	12.0	4.0	21.3	23.3	0.0	0.0	33.3
Flatulence	65	33.8	27.2	41.0	29.8	33.3	0.0	33.3
Fecal Incontinence	63	25.4	18.5	32.8	28.5	33.3	0.0	33.3
Sore Skin	63	26.5	19.0	34.4	31.2	33.3	0.0	33.3
Stool Frequency	62	31.5	25.0	38.7	27.2	33.3	4.2	50.0
Embarrassment	63	31.7	22.8	41.8	38.1	0.0	0.0	66.7

Comparison with reference values

Displays the respective mean values together with published reference data from the German general population, standardized for age and sex distribution, and the EORTC validation cohort of CRC patients after curative-intent treatment. The weighting of the German reference values by age and sex is detailed in Supplement S6. Supplement S7 provides a comprehensive overview of the mean value comparisons shown in Fig. 1. Overall, the mean HRQoL values were lower compared with the German general population’s functioning scores (d = − 0.27 to d = − 0.95). Similarly, mean symptom scores were higher in our cohort (d = 0.08 to d = 0.80). We found a small effect regarding the “Global Health Scale” (− 5.2 points, d = − 0.27). The strongest effects (d ≥ 0.8) were observed for role functioning (− 28.7 points, d = − 0.95), social functioning (− 25.3 points, d = − 0.89), and diarrhea (+ 18.7 points, d = 0.80). Of note, mean scores for financial difficulties were higher in the TNT cohort than in the German general population (+ 12.8 points, d = 0.51). Only a marginal difference (d ≤ 0.2) was found for “Pain” (− 2.1 points, d = − 0.08). The EORTC CRC-patient population had comparable QLQ-C30 function scores, with differences in mean values being < 10 points (d < 0.2). Regarding the three available symptom scores, we found lower mean values in our cohort for Fatigue and Nausea/Vomiting (− 12.0 points, d = − 0.30 and − 4.0 points, d = − 0.26, respectively). Comparing the QLQ-CR29 values, we found comparable function scores, but mean sexual interest was higher in our cohort (men: + 19.4 points, d = 0.62; women: + 10.2 points, d = 0.43). Most of the mean symptom scores were slightly higher in our cohort (< 10 points), except for “Fecal Incontinence” (+ 13.4 points, d = 0.52), “Embarrassment” (+ 12.7 points, d = 0.37),”Impotence” in men (+ 29.0 points, d = 0.73) and “Dyspareunia” in women (+ 10.4 points, d = 0.40). In contrast, “Abdominal Pain”, “Dry Mouth”, and “Stoma Care Problems” mean values were slightly lower (− 4.0 to − 9.1 points; d = − 0.18 to − 0.34).

Fig. 1.

Mean values of quality of life (HRQoL) of the TNTox cohort, EORTC reference values for colorectal cancer (CRC) patients after curative-intent treatment (Whistance et al. 2009), and reference values for the German general population for QLQ-C30 (A + B) and QLQ-CR29 (C–E) subscales. QLQ-CR29 Bowel Symptoms are stratified per presence of a stoma for the TNTox cohort and for the CRC cohort. Higher values in the functioning scales (A + C) and lower values in the symptom scales (B, D, E) indicate better HRQoL in the respective subscales. *No values were available for the CRC cohort in these subdomains

Subgroup and correlation analyses

We conducted a comparison of the above-defined scales between patients with and without colostomy as well as patients with or without NOM, which is shown in Fig. 2. For both groups, no significant differences were observed, except for “Defecation/Stoma-related Problems” (QLQ-CR29), where stoma patients showed lower ranks (p = 0.034, not significant after Bonferroni correction; Fig. 1 and Supplement S4). Furthermore, ranks were not significantly different between men and women. Of note, significantly worse global HRQoL (according to the “Global Health Scale”, p = 0.002, and the “Summary Score”, p < 0.001) as well as higher “Defecation/Stoma-related Problems” (p = 0.003) were observed in patients presenting with symptomatic chronic toxicity, i.e., CTCAE grade ≥ 2. HRQoL ranks were not significantly different in patients who were diagnosed with recurrent disease at follow-up. Detailed test statistics for these known group comparisons are provided in Supplement S8. Furthermore, we found a significant correlation of patients’ age with “Body Image” (Spearman’s ρ = 0.253, p = 0.036). The duration of TNT inversely correlated with the “Global Health Scale” (Spearman’s ρ = − 0.251, p = 0.037) as well as the QLQ-C30 “Summary Score” (Spearman’s ρ = − 0.276, p = 0.027), all not significant after Bonferroni correction. For these scales, a positive correlation with KPS at follow-up was observed (Spearman’s ρ = 0.495, p < 0.001 and ρ = 0.436, p < 0.001, respectively). Additional correlation analyses are reported in Supplement S9.

Fig. 2.

Boxplots of relevant multi-item scales, including the QLQ-C30 “Global Health Status” and “Summary Score”, as well as the QLQ-CR29 scales “Defecation/Stoma-related Problems” and “Body Image”, for known-group comparisons (Watch and Wait, A–D; colostomy, E–H). “Watch and Wait” refers to patients who did not undergo radical surgery at the time of follow-up

Discussion

We analyzed the health-related HRQoL following TNT in a prospective multicenter cohort in Germany. Our data showed a relevant reduction of HRQoL in most domains in comparison with an age- and sex-corrected cohort of a large prospective reference group from the German general population (N = 2448). However, the HRQoL appeared to be comparable with the initial EORTC validation cohort of CRC patients. These results are of particular interest, as certain HRQoL domains were impaired after rectal cancer treatment in several studies (Giandomenico et al. 2015). TNT is linked to a substantial improvement in clinical outcomes, such as overall survival (Conroy et al. 2021) or increased CR rates (Donnelly et al. 2023; Turri et al. 2024). Some earlier trials have demonstrated an increase of toxicity relative to standard CRT or SCRT, which predominantly involves hematological side effects and bowel-related symptoms, specifically diarrhea (Turri et al. 2024; Seow et al. 2025). Moreover, there have been reports of elevated incidences of fatigue or neurotoxicity, i.e., PNP. However, overall toxicity and treatment-related mortality were not significantly different between standard CRT and different TNT regimes in a network meta-analysis (Seow et al. 2025). Consistent with these reports, we did not observe differences in general HRQoL compared to the EORTC reference cohort, which included CRC patients after different treatments with curative intent. Given the substantial number of long-term cancer survivors after curative-intent treatment of LARC, these findings may be of clinical relevance for treatment planning and shared decision making, as they suggest that the oncological benefits of TNT do not necessarily come at the cost of HRQoL.

Compared to the general population, several studies reported a significant reduction in specific HRQoL domains, although general QoL appeared not to be reduced in a review (Giandomenico et al. 2015). In contrast, we found both reduced global and functional QoL compared to the general population. This observation may not only relate to our primary data but also to the selection of the reference group. We referred to the reference values provided by Hinz et al. (2014), who collected their questionnaires from participants contacted randomly across Germany in face-to-face interviews. Many other studies included in that review obtained their normative data from respondents of internet-based questionnaires, making these samples susceptible to sampling bias (Basch et al. 2018). For example, Nolte et al. (2019) reported nearly 10 points lower HRQoL for most QLQ-C30 domains in a comparable reference cohort of the German general population, which was obtained from internet-based interviews. Referring to our values and those presented by Nolte et al., we would also assume less impact on HRQoL. Secondly, age- and sex-dependent differences in HRQoL have been reported (Hinz et al. 2014, 2019; Nolte et al. 2019, 2020), likely influencing the comparison in the case of demographic differences. We mitigated this latter effect by applying age- and sex-weighted reference values. Thus, the selection of appropriate reference cohorts is crucial and might significantly influence the conclusions of such analyses.

There seemed to be no relevant difference in HRQoL across most domains in our patients stratified by NOM and stoma, except for possibly higher bowel-related symptom burden in patients without a stoma. These results are in contrast with the literature; a better bowel function in NOM-patients has been reported (Bach et al. 2021; Quezada-Diaz et al. 2020) as well as a negative impact of a colostomy on HRQoL (Downing et al. 2019; Vonk-Klaassen et al. 2016). Bowel-related symptom scores appeared higher in our cohort compared to the EORTC reference values, which might be a result of the short time interval after TNT in many patients. Whether this observation is further attributed to differences in treatment intensity or to the fact that NOM patients were more likely to suffer from chronic rectal inflammation cannot be concluded. However, caution is advised in interpreting our results due to the small sample sizes in these subgroups. Restorative surgery, i.e., the closure of a stoma after rectal cancer resection, was not found to improve HRQoL in general in a systematic review (Lawday et al. 2022). In contrast, another review reported reduced HRQoL in patients with stoma-related problems (Vonk-Klaassen et al. 2016). We also found lower ranks of most of the multi-item scores for patients with symptomatic chronic toxicity and, inversely, better HRQoL with increased KPS at follow-up. Although guidelines for the reporting of patient-reported outcomes (PROs), such as PRO-CONSORT or ISOQOL, have been published, a review found substantial inconsistency between clinician and patient reporting of toxicity (as measured with HRQoL) for rectal cancer (Gilbert et al. 2015).

In our cohort, a high grade of erectile dysfunction was reported along with high HRQoL symptom scores in men. Despite the small sample size, these impairments seemed to be more common in men than in women. Impairment of sexual function after rectal cancer treatment is frequently described in the literature for both men and women (Ho et al. 2011; Lim et al. 2014). However, its severity seemed to exceed the values provided by the EORTC CRC cohort. A high cumulative oxaliplatin dose was administered to the patients of our cohort, which led to symptomatic PNP in more than one out of four patients. Oxaliplatin may also cause impairment of vegetative functions, especially erectile function (Loprinzi et al. 2020; Cersosimo 2005). Whether this explains the increased symptom scores and why a relatively lower incidence of toxicity was reported for women cannot be concluded from this analysis. HRQoL data from the PROSPECT trial, in which neoadjuvant CRT was compared with neoadjuvant chemotherapy (FOLFOX), indicate that patients treated with chemotherapy alone experienced less diarrhea and better bowel function than those receiving CRT, but a higher burden of other symptoms (e.g., fatigue, anxiety, nausea, and neuropathy) (Basch et al. 2023). At 12 months, FOLFOX patients reported lower rates of fatigue, neuropathy, and sexual dysfunction. These findings suggest that the more intensive systemic treatment may be associated with a (transient) impairment of HRQoL, while CRT could have a negative impact on long-term HRQoL. As different QoL questionnaires were used, a direct comparison with our cohort was not possible.

The reported financial difficulty scores were higher than in the general population. This observation is in line with a German sample of patients with different cancer types after radiotherapy. In that cohort, 29% of rectal cancer patients reported financial difficulties related to their treatment (Fabian et al. 2025; Wurschi et al. 2024).

Strengths and limitations

We prospectively collected standardized questionnaires from a multicenter cohort throughout Germany. All patients underwent a comparable treatment sequence, consisting of neoadjuvant RT or CRT, followed by consolidation chemotherapy. The use of the QLQ-C30 and QLQ-CR29 enables comparison with various cohorts; employing age- and sex-weighted reference values from the German general population helped to mitigate the influence of demographic differences. This standardization was not applied to reference values from the EORTC cohort of CRC patients, due to the absence of age and sex subgroups. With a mean age of 65 years and 58% male patients, the demographic influence appeared relatively minor.

However, several limitations must be acknowledged. First, this is a relatively small sample from a non-randomized cohort of patients. HRQoL questionnaires were collected in only 10 of the 23 participating centers, primarily due to local data protection regulations or institutional policies. Consequently, the HRQoL sample does not represent the entire cohort, and the generalizability of the HRQoL findings may be limited. With a median follow-up interval < 18 months, recovery from chronic toxicity, such as fatigue, persistent bowel irritation or neuropathy, may not yet be complete in all patients. This may affect the interpretation of our results in comparison with cohorts with longer follow-up periods (Rausa et al. 2017). Nonetheless, this factor was assessed in exploratory correlation analyses and did not show a significant impact on the multi-item scales in our cohort. As we included patients over a period of nearly one decade, advances in supportive care and their gradual implementation into routine clinical practice must be taken into account. In Germany, the national S3 guideline on supportive care has been available since 2016, and its recommendations may have influenced the management of some patients included in this study. In this context, CRC patients of the EORTC reference group (published in 2009) may not have received current treatment techniques (such as intensity-modulated radiotherapy, IMRT) or the same supportive care measures, which could potentially influence the interpretation of our results.

Due to the small sample size, no complex regression analyses were conducted. Moreover, the cross-sectional approach does not allow for longitudinal comparison or adjustment to baseline values, which would be essential to differentiate between comorbidity-related impairments of HRQoL and treatment-related impairments. Additionally, no information on income or educational status was available for our cohort.

Conclusion

In our cohort of rectal cancer patients, HRQoL was lower following TNT compared with the German general population, but comparable with most HRQoL subscales of CRC patients. HRQoL was similar in most subgroup analyses regarding NOM and the presence of a stoma. However, patients with symptomatic toxicity at follow-up reported lower HRQoL within the QLQ-C30 questionnaire. These findings highlight the importance of professional healthcare management to maintain quality of life among cancer survivors. Future trials may include QoL endpoints to improve rectal cancer treatment from both the clinician’s and the patients’ perspective.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

95% CI

95% Confidence interval

APC

Argon plasma coagulation

ASCRS

American Society of Colon and Rectal Surgeons

CR

Complete response

CRT

Chemoradiotherapy

CRC

Colorectal cancer

CTCAE

Common terminology criteria of adverse events

DEGRO

Deutsche Gesellschaft für Radioonkologie (German Society for Radiation Oncology)

EMVI

Extramural vascular invasion

EORTC

European Organisation for Research and Treatment of Cancer

ESMO

European Society for Medical Oncology

HRQoL

Health-related quality of life

IMRT

Intensity-modulated radiotherapy

KPS

Karnofsky performance status

LCRT

Long-course chemoradiotherapy

MRF

Mesorectal fascia

NOM

Non-operative management

PRO

Patient-reported outcome

QoL

Quality of life

SCRT

Short-course radiotherapy

TNT

Total neoadjuvant therapy

Author contributions

Georg Wurschi: Conceptualization, Investigation, Data curation, Methodology, Formal analysis, Investigation, Methodology, Project administration, Visualization, Writing—original draft, Writing—review & editing. Melanie Schneider: Investigation, Writing—review & editing. Jan-Niklas Becker: Investigation, Writing—review & editing. Felix Ehret: Investigation, Writing—review & editing. Markus Diefenhardt: Investigation, Writing—review & editing. Alexander Rühle: Conceptualization, Investigation, Writing—review & editing. Daphne Schepers von Ohlen: Investigation, Writing—review & editing. Justus Kaufmann: Investigation, Writing—review & editing. Alina Depardon: Investigation, Writing—review & editing. Hai Minh Ha: Investigation, Writing—review & editing. Adrianna Cieslak: Investigation, Writing—review & editing. Maren Schöneich: Investigation, Writing—review & editing. Florian Rißner: Resources, Project administration, Writing—review & editing. Maximilian Römer: Investigation, Writing—review & editing. Andreas Hinz: Methodology, Writing—review & editing. Klaus Pietschmann: Conceptualization, Supervision, Resources, Writing—review & editing.

Funding

Open Access funding enabled and organized by Projekt DEAL. No funding was received for conducting this study.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing Interests

Felix Ehret has received honoraria and travel support from ZAP Surgical Systems, Inc. and Accuray, Inc., and acknowledges research funding from the German Cancer Aid and Accuray, Inc., all unrelated to the submitted work. All other authors report having no conflicts of interest regarding the submitted work.

Ethics approval

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Jena University Hospital (2023-3042-Bef).

Consent to participate

Informed consent was obtained from all individual participants included in the study.