Patient-Reported Outcomes After Neoadjuvant Therapy and Watch-and-Wait for Rectal Cancer: A Systematic Review and Meta-Analysis

Abstract

Objective:

To evaluate patient-reported outcomes (PROs) in rectal cancer patients who underwent neoadjuvant therapy followed by watch-and-wait.

Background:

Advancements in rectal cancer therapies have improved oncologic outcomes and increased focus on long-term treatment effects. Watch-and-wait, which aims for organ preservation in patients with a clinical complete response to neoadjuvant therapy, is used more frequently. While watch-and-wait may reduce functional impairments associated with proctectomy, its association with PROs is poorly understood.

Methods:

We performed a systematic review and meta-analysis of studies reporting PROs for patients with stage I–III rectal adenocarcinoma treated with neoadjuvant therapy and watch-and-wait. Random-effects model meta-analysis was performed when feasible; otherwise, data were synthesized through a narrative review.

Results:

Twenty-five studies (3139 patients) reported PROs for patients receiving neoadjuvant therapy followed by watch-and-wait (n = 1283), proctectomy (n = 1583), local excision (n = 263), or declined surgery (n = 10). Most studies (64.0%) excluded patients with local regrowth, leaving PRO data for 103 such patients. Fourteen studies compared PROs between treatment groups after neoadjuvant therapy. Meta-analyses demonstrated fewer bowel symptoms and improved quality of life (QoL) in 1/6 subscales for watch-and-wait. Across 24 QoL comparisons, 14 showed improvement in some subscales for watch-and-wait, while 10 found no difference. For 24 symptom scale comparisons, 11 indicated some improvement with watch-and-wait, and 13 found no differences between groups.

Conclusions:

Watch-and-wait offers the potential to avoid permanent ostomy and surgical morbidity. However, more robust data are needed to assess its association with PROs compared to proctectomy or local excision. Researchers should prioritize publishing QoL data to better inform shared decision-making.

INTRODUCTION

Rectal cancer is a significant global health issue, with over 730,000 cases diagnosed worldwide in 2020 and more than 46,000 anticipated in the United States in 2024.^1,2 Recent advancements in treatment and regulatory agencies’ emphasis on quality of life (QoL),^3,4 have increased attention to the long-term effects of cancer treatment. As therapeutic options for rectal cancer increase, with several approaches offering comparable oncologic outcomes, the impact of treatment on a patient’s physical well-being, emotional health, and functional status becomes increasingly important.

Patient-reported outcomes (PROs) are data collected directly from patients about their health across several interrelated domains including global QoL, condition- or disease-specific QoL, and symptom scales, each serving a distinct purpose.⁵ Global QoL measures provide a broad assessment of overall well-being, while condition-specific QoL instruments focus on the unique impact of a particular disease or treatment. Symptom scales capture specific physical or emotional challenges patients experience, which may or may not correlate with QoL.^6–9 Understanding PROs, including both symptom and QoL metrics, helps clinicians and patients select therapies that not only aim to cure cancer but also minimize side effects and improve overall patient satisfaction. Proctectomy, while potentially curative, often requires a permanent ostomy, which can profoundly impact daily life.^10–13 Nonetheless, most treatments affect PROs, including not only surgery but also radiation and chemotherapy.^9,14–18

The increasing use of watch-and-wait for select patients with rectal cancer following neoadjuvant therapy¹⁹ may offer an opportunity to reduce some surgery-related impairments, with preliminary studies suggesting improved QoL compared to proctectomy.^20–22 However, although this treatment strategy was first proposed as a viable alternative to proctectomy for select patients more than 20 years ago,²³ no systematic synthesis of PROs has been conducted to fully understand its impact. We hypothesized that existing data would be heterogeneous, but that watch-and-wait would be associated with improved QoL and fewer symptoms compared with proctectomy or local excision (LE).

METHODS

This systematic review was preregistered on PROSPERO (CRD42023450775) and follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and Meta-analyses of Observational Studies in Epidemiology guidelines (Tables S1 and S2, https://links.lww.com/AOSO/A525, protocol amendments in Table S3, https://links.lww.com/AOSO/A525).²⁴ Research board approval was not required as this is a secondary analysis of published data.

Search Strategy

With the assistance of an experienced medical librarian, we searched PubMed, Embase, and Web of Science through October 1, 2024 without language restrictions. We supplemented our searches by citation tracking and a gray literature search on Google Scholar on October 30, 2024,²⁵ review of the bibliographies of included studies, and a search of studies on ClinicalTrials.gov was performed on March 12, 2024. Lastly, an individual with expertise in the field (FF) was consulted regarding any missing studies. The searches are in Table S4, https://links.lww.com/AOSO/A525.

Study Selection

Four reviewers, 2 per study (B.H.C., A.L., T.D., and L.R.), independently reviewed all citations for eligibility using Covidence software (Veritas Health Innovation, Melbourne, Australia). We included studies with PROs for patients with stage I–III rectal adenocarcinoma who underwent neoadjuvant therapy and watch-and-wait. We excluded studies without PROs, without neoadjuvant therapy, without patients undergoing watch-and-wait, those with <10 patients, and editorials, letters to the editor, or studies without original data. Given the limited number of randomized controlled trials (RCTs) that have reported PROs, we included any available RCTs in addition to nonrandomized studies of interventions (NRSI) and cohort studies. Qualitative studies were not included since these require different methodologies to synthesize.

Data Extraction

Using a standardized data extraction form, 4 reviewers (2 per study; B.H.C., A.L., T.D., and L.R.) collected data from eligible studies. Abstracted variables of interest are presented in Table S5, https://links.lww.com/AOSO/A525, with funding sources for each study presented in Table S6, https://links.lww.com/AOSO/A525. For studies that captured PRO data at multiple timepoints, the timepoint with the longest follow-up was selected for inclusion in the meta-analysis to assess longest-term PROs and the time at collection is presented along with the results in Tables S7 and S8, https://links.lww.com/AOSO/A525. PRO instruments are categorized into generic QoL, condition or disease-specific QoL, and symptom scales, based on the intended population and question content, with PRO instruments used by included studies outlined in Table 1. The characteristics of each PRO instrument are in Table S9, https://links.lww.com/AOSO/A525.

TABLE 1.

Quality of Life (QoL) and Symptom Scale Instruments Used in Included Studies

Type of QoL Instrument	QoL Instrument Used	Studies
Generic QoL	SF 36	Custers et al,22 Hupkens et al26
Cancer-specific QoL	EORTC-QLQ-C30	Geubels et al,27 Pennings et al,28 Custers et al,22 Bach et al,29 Bulens et al,30 Reddy et al,31 Wang et al,32 Chiloiro et al,33 Custers et al,34 Jones et al,35 Hupkens et al26
	FACT-G7	Rooney et al36
Disease-specific QoL	EORTC-QLQ-CR29/38	Geubels et al,27 Pennings et al,28 Custers et al,22 Reddy et al,31 Wang et al,32 Chiloiro et al,33 Custers et al,34 Dizdarevic et al,37 Jones et al,35 Hupkens et al,26 Appelt et al38
	FIQoL	Rooney et al,36 Kaul et al,39 Chiloiro et al,33 Pascual-Russo et al,40 Habr-Gama et al20
	FACT-C	Kaul et al,39 Kim et al41
	EQ-5D-3L	Jones et al35
Type of Symptom	Symptom Scale Used	Studies
Bowel function	LARS	Geubels et al,27 Pennings et al,28 Custers et al,22 Rooney et al,36 Bulens et al,30 Reddy et al,31 Custers et al,34 Faaborg et al,42 Haak et al,43 Jones et al,35 Van der Sande et al,44 Dhadda et al,45 Hupkens et al26
	Vaizey	Custers et al,22 Custers et al,34 Haak et al,43 Jones et al,35 Van der Sande et al,44 Hupkens et al,26 Martens et al46
	Wexner	Faaborg et al,42 Pascual-Russo et al,40 Habr-Gama et al,20 Maas et al47
	MSK-BFI	Bulens et al,30 Quezada-Diaz et al,21 Maas et al47
Urinary function	IPSS	Custers et al,22 Haak et al,43 Jones et al,35 Hupkens et al26
Sexual function	IIEF	Custers et al,22 Bulens et al,30 Jones et al,35 Hupkens et al26
	FSFI	Custers et al,22 Bulens et al,30 Hupkens et al26
	MFSQ	Jones et al35

EORTC-QLQ-C30/CR29 or 38 indicates European Organization for Research and Treatment of Cancer-Quality of Life Questionnaire Core 30/Colorectal 29 or 38; EQ-5D-3L, European Quality of Life 5 Dimensions 3-Level; FACT-G/C, functional assessment of cancer therapy general/colorectal; FIQoL, fecal incontinence quality of life; FSFI, Female Sexual Function Index; IIEF, International Index of Erectile Function; IPSS, International Prostate Symptom Score; LARS, low anterior resection syndrome score; MFSQ, McCoy Female Sexuality Questionnaire; MSK-BFI, Memorial Sloan Kettering Bowel Function Index; QoL, quality of life; SF-36, short form health survey 36; Vaizey, Vaizey Incontinence Score; Wexner, Wexner Incontinence Score.

Risk-of-Bias Assessment

Two authors (M.B. and M.J.) conducted an independent and duplicate assessment of bias risk using the Risk of Bias in Nonrandomized Studies of Interventions framework for RCT and NRSI and the Risk of Bias in Nonrandomized Studies of Exposures tool for studies without a comparator group.^48,49 Discrepancies were discussed to achieve consensus with biases graded per the guidelines for each tool.

Statistical Analysis

A meta-analysis comparing treatment groups was possible as required data were available from >1 study for 2 disease-specific QoL instruments [EORTC-QLQ-C30, Fecal Incontinence Quality of Life (FIQoL)] and 1 symptom scale [Low Anterior Resection Syndrome (LARS) Score]. The single RCT was excluded from meta-analysis as all other studies were NRSIs. Estimates of pooled treatment effect (such as watch-and-wait or surgery) for each PRO instrument or subscale were performed using the sample size, mean, and standard deviation (SD), with comparisons between treatment groups using standardized mean differences (SMD) and standard error. For studies that reported proportions or alternative treatment effect measures such as median and range, standardized methods were employed to derive the estimated means and SD.⁵⁰ In cases where these values were not provided, corresponding authors were contacted via email to request the necessary data. SMD was selected to compare treatment effects over comparing minimal clinically important differences (MCID) to allow for appropriate weighting of effect by study. A random-effects model with the restricted maximum likelihood estimator⁵¹ and the Knapp and Hartung⁵² adjustment was used to analyze pooled outcomes, estimating the mean and 95% confidence interval (CI) for each QoL and functional assessment in patients on watch-and-wait. For studies with a comparator group, SMDs and 95% CI between treatments were calculated. High heterogeneity, as measured by I², was investigated further by analysis of potential subject or subgroup differences that could explain the results. Analysis was performed using the meta package in RStudio v 2024.09.1 (Boston, MA).

Due to the variety of tools used to assess PROs, inconsistent data reporting, and the absence of a comparator group in some studies, pooling treatment effects was not always possible. As a result, a narrative review was conducted to summarize these findings. The main text highlights results for the most commonly used instruments across studies, while findings from other instruments are tabulated, with QoL instruments in Table S7, https://links.lww.com/AOSO/A525 and symptom scales presented in Table S8, https://links.lww.com/AOSO/A525. For instruments with statistically significant differences between treatment groups which could not otherwise be pooled either due to missing information to calculate variance, required for meta-analysis, or because numerical data were available for only 1 study, absolute differences in mean between groups were calculated, compared to established MCIDs, and presented graphically.⁵³

RESULTS

A total of 5262 studies were identified, with 4658 remaining for screening after duplicate removal, 4438 were excluded based on irrelevance and 221 full-text articles were reviewed (Fig. 1). Ultimately, 25 studies were included (Table S5, https://links.lww.com/AOSO/A525). The majority (72.0%) originated from Europe, were prospective (68.0%), were conducted at a single institution (72.0%), and either had no alternative treatment group for comparison (40.0%) or compared watch-and-wait with proctectomy after neoadjuvant therapy (36.0%, Table 2). Notably, 32.0% (n = 8) of the studies including 746 patients, shared overlapping patient populations, and 64.0% (n = 16) of the studies excluded watch-and-wait patients who experienced local regrowth, leaving PRO data for 103 patients on watch-and-wait with local regrowth.

FIGURE 1.

Study selection. PRISMA diagram of article selection.

TABLE 2.

Summary of Characteristics and Patients in Included Studies

Variable, No. (%)	All Included Studies/Patients
Total number of included studies	n = 25
Geographic location
Asia	1 (4·0)
Europe	18 (72·0)
North America	4 (16·0)
South America	2 (8·0)
Year of study
2011–2015	2 (8·0)
2016–2020	9 (36·0)
2020–2024	14 (56·0)
prospective study	17 (68·0)
Multicenter study	7 (28·0)
Comparator group
Healthy controls	1 (4·0)
Local excision (TEMS, TAMIS)	3 (12·0)
Local excision and proctectomy	3 (12·0)
None	10 (40·0)
Proctectomy	9 (36·0)
Total number of included patients	3139
Clinical stage
AJCC stage	1335 (42·5)
Stage 1	300 (22·5)
Stage 2	251 (18·8)
Stage 3	784 (58·7)
TNM classification	1804 (57·5)
T1–2	453 (25·1)
T3–4	1127 (62·5)
Unknown T	224 (12·4)
N0	644 (35·7)
N+	976 (54·1)
Unknown N	184 (10·2)
Known tumor distance from anal verge (cm)	1409 (44·9)
Categorical distance, cm	302 (21·4)
<5	215 (71·2)
<6	13 (4·3)
5–10	67 (22·2)
>10	7 (2·3)
Mean distance, n = 82 (5·8%)	2·7 (SD 1·8)
Median distance, n = 1025 (72·7%)	4·1 (Range 0–15)

TAMIS indicates transanal minimally invasive surgery; TEMS, transanal endoscopic microsurgery.

Patient and Treatment Characteristics

Identifying the precise number of unique patients is challenging due to the likelihood of participant overlap. Among the 746 potential duplicated participants, specific identification was not possible. Consequently, the total cohort of 3139 patients likely includes some duplicates (Table 2). Of this cohort, 44.9% (n = 1409) had a reported tumor distance from the anal verge, with a median distance of 4.1 cm presented in 72.7% of cases (n = 1025), a mean distance of 2.7 cm presented in 5.8% of cases (n = 82), and categorical distance presented in 21.4% (n = 302), with a total of 215 of these patients (71.2%) having tumors within 5 cm of the anal verge (Table 2). Treatment regimens included long-course chemoradiotherapy (47.9%, n = 1503), short-course radiotherapy (18.6%, n = 584), total neoadjuvant therapy (3.3%, n = 105), or a radiation-sparing regimen (proctectomy or LE alone; 2.5% n = 80), among other combinations. A small subset of patients (4.0%, n = 126) received multiple radiotherapy types to include short-course radiotherapy or long-course chemoradiotherapy with contact brachytherapy. Of the included patients, 49.3% (n = 1546) received organ-preserving strategies: 40.9% (n = 1283) via watch-and-wait and 8.4% (n = 263) with LE. PRO data were inconsistently captured across studies. Some excluded patients with permanent ostomies from reporting PROs (eg, LARS Scores), and variable follow-up durations within studies made it difficult to determine the precise number of patients contributing data at each time point.

Meta-Analyses

A meta-analysis comparing treatment groups was possible for 2 disease-specific QoL instruments (EORTC-QLQ-C30, FIQoL) and 1 symptom scale (LARS Score). Comparing watch-and-wait to proctectomy for EORTC-QLQ-C30 in 3 studies with available data, physical functioning for watch-and-wait demonstrated significant improvement over proctectomy, but global health status, emotional and social functioning, fatigue, and pain were not consistently significantly different (Fig. 2A). The meta-analysis of the FIQoL instrument demonstrated no significant difference between watch-and-wait and proctectomy in the 3 studies with complete available data, though it was not the most common instrument assessing disease-specific QoL thus results appear in the supplement (Figure S1A, https://links.lww.com/AOSO/A525).

FIGURE 2.

Results of meta-analyses. Forest plots from random-effects model for select EORTC-QLQ-C30 (A) and LARS (B) scores.

Four studies with complete data demonstrated improved LARS scores for watch-and-wait as compared to both LE and proctectomy, though there was insufficient evidence when comparing each group individually (Fig. 2B). Pooled estimate of LARS score after watch-and-wait for ten studies with available data demonstrated that with this treatment strategy, patients generally have “minor LARS” as defined by scoring guidelines for this instrument, with a pooled LARS score of 22.0 (95% CI = 20.7–23.4, Figure S1B, https://links.lww.com/AOSO/A525).

High heterogeneity was noted in the meta-analysis of the FIQoL instrument, and the EORTC-QLQ-C30 global health score, fatigue, and pain scores. In comparison of the studies used for EORTC-QLQ-C30, 1 had PRO data for only 29 patients as compared to over 100 in the other 2 studies, which may suggest that they had insufficient power to detect a difference between treatment groups. Moreover, evaluation of the pooled estimate of scores for watch-and-wait across instruments demonstrated less than 55%, or negligible to moderate heterogeneity for all subscales (Figure S1, https://links.lww.com/AOSO/A525).

Narrative Review

Generic Quality of Life

Three studies report generic QoL data for patients who chose watch-and-wait (Table S5, https://links.lww.com/AOSO/A525). Two found similar QoL outcomes between watch-and-wait and LE, 1 using the SF-36 and the other EQ-5D-3L instrument (Table S7, https://links.lww.com/AOSO/A525). However, in 2 studies comparing watch-and-wait to proctectomy using the SF-36, the watch-and-wait group consistently showed significant improvements in multiple subscales (Fig. 3). One of these studies found that patients after proctectomy reported better general health on that subscale as compared to watch-and-wait (Fig. 3). Overall, generic QoL outcomes suggest watch-and-wait is comparable to LE and offers advantages over proctectomy, especially in physical function and physical role functioning.

FIGURE 3.

Differences in other QoL or function scales between treatment groups. Absolute difference in each measure reported by studies that demonstrated statistically significant differences between the watch-and-wait and comparator group, though with insufficient data for pooled analysis of treatment effect between groups. A indicates anxiety; AP, abdominal pain; BFI, Memorial Sloan Kettering Bowel Function Instrument; BI, body image; BM, blood & mucous in stool; BP, buttock pain; CCS, FACT-C colorectal cancer subscale; DM, dry mouth; DP, defecation problems; E, embarrassment; ERF, emotional role function; GH, general health; MH, mental health; P, pain; PF, physical function; PRF, physical role function; SF, sexual function; T, taste; TOI, FACT-C Trial outcome index; UF, urinary frequency; UI, urinary incontinence; V, vitality (energy/fatigue).

Disease or Cancer Specific Quality of Life

Eighteen studies assessed cancer-specific or colorectal cancer-specific QoL (Table S5, https://links.lww.com/AOSO/A525). The EORTC-QLQ-C30 was used by 11 studies, 7 of which had comparator groups including 159 patients who underwent neoadjuvant therapy and proctectomy, 65 with neoadjuvant therapy and LE, and 40 with LE alone (Table S7, https://links.lww.com/AOSO/A525). One study found improvement only in nausea for watch-and-wait as compared to LE after neoadjuvant therapy, whereas 2 found no differences between these groups. In comparison to proctectomy, studies found significant improvements for watch-and-wait in global health status (2 studies), physical functioning (2 studies), financial difficulty (2 studies), pain (2 studies), role, cognitive, emotional and social functioning, global health, fatigue, pain, financial difficulty, appetite loss, and dyspnea all exceeding MCID aside from dyspnea where scores were not provided to evaluate. One study found no difference in total score between groups, 1 found no difference in subscales or total score between watch-and-wait or proctectomy, and 1 study did not report the results of this instrument across groups. Overall, watch-and-wait was comparable to LE, though showed advantages in specific domains including financial difficulties and some functional or symptom subscales such as physical functioning, compared to proctectomy.

The EORTC-QLQ-CR29 and -CR38 were the most used colorectal cancer-specific instruments, utilized in 11 studies (one using -CR38 exclusively and one using both -CR38 and -CR29). Five studies had comparator groups including 126 patients who underwent neoadjuvant therapy and proctectomy, 65 with neoadjuvant therapy and LE, and 40 with LE alone (Table S7, https://links.lww.com/AOSO/A525). Comparing watch-and-wait with LE, 1 study found improved anxiety whereas 1 found less embarrassment and less blood/mucous in stool for watch-and-wait, all exceeding MCIDs (Fig. 3). In comparison to proctectomy, studies found improved body image (3 studies), sexual function (2 studies), urinary frequency and incontinence, defecation problems, abdominal and buttock pain, dry mouth, and taste for watch-and-wait though only body image, sexual function, and urinary frequency exceeded MCIDs, aside from body image in 1 study as values were not provided (Fig. 3). One study did not report the results of this instrument across groups. Overall, watch-and-wait demonstrated advantages in specific domains such as body image, sexual functioning, embarrassment, and in anxiety, urinary frequency, and bloody stool.

Symptom Scales – Bowel Function

Bowel function was assessed in 18 studies, most frequently with the LARS score (Table S5, https://links.lww.com/AOSO/A525). The LARS score was used in 12 studies, with 7 studies including a comparator group comprised of 126 patients who underwent proctectomy, 44 patients who underwent LE after neoadjuvant therapy, 40 patients who underwent LE alone, and 15 normal controls (Table S8, https://links.lww.com/AOSO/A525). Two studies comparing watch-and-wait to LE after neoadjuvant therapy found significantly lower LARS scores, indicating fewer symptoms, in the watch-and-wait group. Of the 4 studies comparing watch-and-wait to proctectomy without permanent ostomy, 2 found a significantly lower in LARS score for watch-and-wait compared to proctectomy, and 2 did not find a difference.

Symptom Scales – Urinary and Sexual Function

Urinary function was evaluated in 4 studies, with no differences found when comparing watch-and-wait to LE, and 2 studies finding some improvement in urinary function for watch-and-wait as compared to proctectomy (Table S8, https://links.lww.com/AOSO/A525). Sexual function was assessed in 4 studies. One study had too few data to report on female sexual function, and the other 3 found no differences in sexual function by treatment group. For erectile function, there were no differences noted between groups across all 4 studies.

Risk of Bias

Two of the studies had a critical risk of bias due to selective reporting of results (Figure S2, https://links.lww.com/AOSO/A525). To evaluate for potential publication bias, we included a gray literature search. Given the limited number of studies with comparisons and numerical data to calculate SMD between treatments, a funnel plot would not be insightful. However, reviewing the published results for QoL and symptom scales for patients after watch-and-wait, while most instruments used demonstrated a mixed effect of watch-and-wait as compared to other treatments, studies with positive results were more likely to publish the raw data necessary for a meta-analysis to be performed, which could affect these results. Nonetheless, narrative review supplements the meta-analysis data to offer a more balanced evaluation of the results.

DISCUSSION

To our knowledge, this is the first systematic review and meta-analysis of PROs in rectal cancer patients managed with watch-and-wait. Our findings indicate that generic QoL is largely comparable between watch-and-wait and LE, but watch-and-wait shows clear advantages over proctectomy, particularly in physical and physical role functioning. Furthermore, watch-and-wait generally demonstrates benefits over both proctectomy and LE in disease-specific QoL domains, including bowel function, body image, and financial difficulties, although these results were not uniform across all measures. Finally, while bowel function was generally better with watch-and-wait, other symptom scales, such as sexual and urinary function, did not consistently show benefits compared to proctectomy or LE.

A primary goal of watch-and-wait is to maximize PROs while maintaining oncologic outcomes comparable to proctectomy for patients with locally advanced rectal cancer. However, the literature has largely focused on oncologic outcomes, with less attention to overall patient well-being. While some generic QoL instruments show clear advantages for watch-and-wait over proctectomy, our findings emphasize the inconsistency of benefit across generic QoL measures when compared to other treatment strategies. This is important as cost-comparison analyses often rely on generic QoL instruments, like the EQ-5D, to estimate QoL costs, or savings, for each treatment strategy.^54,55 Current cost analyses generally favor watch-and-wait compared with proctectomy, but they often rely on limited health-related QoL,⁵⁶ sometimes using estimates from prostate cancer, patient preferences instead of QoL data, or omitting QoL altogether.^57–59 One study mapped EORTC QLQ-C30 scores onto the EQ-5D to estimate QoL implications,⁶⁰ however, this approach likely does not accurately reflect QoL gains given the minimal QoL benefit reflected in the generic QoL scales.

Critically, most studies evaluating PROs excluded patients with tumor regrowth, an important issue given that the reduced likelihood of permanent stoma formation likely underlies many of its observed QoL advantages over proctectomy. This QoL benefit related to stoma avoidance is particularly pronounced for distal rectal tumors, where permanent ostomy is more likely after proctectomy. In contrast, sphincter-preserving proctectomy is often feasible for proximal tumors, potentially diminishing the QoL difference between treatment strategies. These factors highlight the pressing need for more nuanced QoL assessments that account for tumor location and regrowth to accurately evaluate the true benefits of watch-and-wait.

Watch-and-wait demonstrates a more consistent benefit over proctectomy and LE for disease-specific QoL. Although global QoL total scores frequently showed no difference, watch-and-wait was linked to better outcomes in multiple subscales (physical, emotional, social, and cognitive) and functional measures within these QoL instruments (pain, fatigue, bowel, sexual, and urinary function). Notably, 1 study comparing watch-and-wait to LE without chemoradiation found that LE demonstrated improved QoL and function across multiple domains.³⁵ This intriguing result suggests that current research exploring de-escalated chemoradiotherapy regimens might lead to even greater improvements in PROs, a hypothesis that requires further investigation.

With regard to the symptom scales, the data remain heterogeneous. Urinary function was improved in some studies for watch-and-wait, though sexual function, as measured by dedicated instruments, appeared similar across treatment groups. However, multiple studies commented on higher percentages of missing data for sexual function, which may be impacting these results. The meta-analyses comparing LARS scores between patients who underwent watch-and-wait versus proctectomy or versus LE involved only 2 studies each, limiting the power to detect differences and hence resulted in estimates of effect with CIs that crossed zero. While proctectomy and LE likely have different effects on long-term bowel function, which we did not aim to characterize, meta-analysis suggests that patients who undergo neoadjuvant therapy followed by either proctectomy or LE experience more LARS symptoms than those on watch-and-wait. Across all included studies, watch-and-wait consistently offered benefit for bowel function as measured by the LARS score, in comparison to proctectomy and LE, but remains significantly worse than bowel function for healthy controls, as would be anticipated. However, given that the LARS instrument was designed for assessing bowel function in patients with and without radiation after proctectomy, there may be a degree of measurement error impacting these results as this was not the intended population for this tool. Other bowel function instruments, including the Vaizey score, Memorial Sloan Kettering Bowel Function Instrument, and Wexner score, inconsistently demonstrated benefit for watch-and-wait, though again these tools were also not specifically intended for this population. Moving forward, patients would benefit from a tool specifically designed to evaluate bowel function for patients on watch-and-wait as compared to other treatment strategies, which captures meaningful differences in bowel function that impact QoL.

Several limitations must be considered when interpreting these findings. First, the PRO instruments used, the neoadjuvant therapies delivered, likely overlapping patient populations, mixed reporting of results, and varying timepoints of assessment make it challenging to aggregate and interpret results. Second, studies underrepresent patients with local regrowth or distant metastasis, possibly leading to an overestimation of the PRO benefits of watch-and-wait. Third, there are no validated instruments for assessing bowel function in the context of watch-and-wait; for example, the LARS score was developed for understanding bowel function in patients with and without radiation after proctectomy though has not been found to reliably correlate with QoL. Finally, contemporary and emerging treatment strategies like total neoadjuvant therapy or radiation-sparing regimens are underrepresented in the literature. These limitations underscore the need for more rigorously designed studies with larger sample sizes and longer follow-up periods to better understand the true impact of watch-and-wait on PROs. Future studies should prioritize PROs as endpoints in clinical trial design, using prespecified, consensus-derived instruments to allow for more comparisons across studies.

For patients and multidisciplinary teams considering watch-and-wait, these findings support claims in the literature that watch-and-wait is associated improvements for some subscales of disease-specific QoL and bowel function measured by the LARS scale, though overall the picture remains mixed such that many PROs on watch-and-wait may be comparable to patients who undergo surgery after neoadjuvant therapy. Researchers should aim to include a broader range of patient experiences, like those with local regrowth and more proximal tumors, and to develop validated bowel function measures specific to watch-and-wait, which capture meaningful bowel function impacting QoL. Policymakers, particularly in countries with single-payer systems, should consider the lack of generic QoL data when evaluating the cost-effectiveness and long-term outcomes of watch-and-wait as a treatment strategy.

In conclusion, PROs in rectal cancer are essential as they will help guide which treatment strategy, neoadjuvant therapy followed by watch-and-wait versus proctectomy, is the first-line recommendation for patients moving forward. However, the field currently lags in both studying and reporting these outcomes, such as QoL and symptom burden, which are important to patients. This review attempts to frame the existing literature, highlighting the inability to draw definitive conclusions about PROs in rectal cancer patients who undergo neoadjuvant therapy and watch-and-wait or rectal resection using current data due to a combination of limited studies and varied instruments used.

ACKNOWLEDGMENTS

Thank you to Linda Hasman, our medical librarian, who was instrumental in developing the search strategy and running the searches to find articles for inclusion in the systematic review.

Supplementary Material

as9-6-e602-s001.pdf ^{(943.3KB, pdf)}