Organ Preservation in Rectal Cancer

Abstract

A proportion of patients with rectal cancer who undergo neoadjuvant treatment end up with a complete pathological response. Because these patients have excellent oncological outcomes, many have questioned if surgical treatment in this population constitutes overtreatment. There is growing interest in organ preservation in rectal cancer patients. We review the fundamentals and key considerations of this approach including patient selection, available treatment regimens, assessment of response, long term monitoring of response, and long term oncological and functional outcomes.

Introduction

The treatment of nonmetastatic locally advanced rectal cancer (LARC) has improved over the last several decades due to advancements in surgical technique, diagnostic imaging, surgical pathology, and perioperative multimodal treatment. Standard treatment for LARC includes preoperative neoadjuvant chemoradiotherapy (CRT), resection with total mesorectal excision (TME), and postoperative adjuvant systemic chemotherapy ¹. Although this aggressive multimodal treatment has been associated with excellent local control, a third of patients still die from metastatic disease ². Multimodal treatment is also associated with morbidity, risk of permanent stoma, and severe functional impairments affecting the patients’ quality of life ^3,4. Key challenges for the next decade are prolongation of survival by preventing distant metastases and improvement in the patients’ quality of life ⁵.

It is well established that a proportion of patients who undergo preoperative CRT end up with a pathological complete response (pCR) ⁶. This proportion can be as high as 44% in highly selected patients with early rectal cancer ⁷. pCR has been observed in 18% of LARC patients who received CRT alone and 38% of LARC patients who received preoperative CRT in combination with systemic chemotherapy ⁸. Patients with pCR have a local-recurrence rate of around 1% and a 5-year survival rate of 95% ^{6, 9}. In view of these impressive outcomes, many have raised the question of whether surgical removal of the rectum in patients with pCR constitutes overtreatment. Avoiding surgery can provide important benefits such as reduced morbidity, lower health care costs, and better quality of life, as long as oncological outcomes are not compromised. Significant progress has been made in exploring watch and wait (WW) with selective rectal preservation as an alternative treatment strategy for LARC.

In deciding on optimal treatment for a patient whose tumor is essentially gone after CRT, the surgeon faces the dilemma of weighing the benefit of avoiding TME against the possibility of the tumor progressing to the point of being no longer curable. The challenge is to correctly identify patients with a complete response who can safely avoid TME.

Acceptance of WW as a viable treatment strategy for LARC is growing, based on encouraging data from retrospective studies conducted mainly at specialized high-volume centers. This article is aimed at clarifying the fundamentals and key considerations of this treatment strategy.

Terminology

pCR is an objective determination that no tumor cells are present in the resected surgical specimen. Clinical complete response (cCR) is a subjective assessment—based on digital rectal examination (DRE), endoscopy with or without biopsy, and MRI—that no viable tumor remains in the rectum following neoadjuvant treatment (NAT). A cCR does not always correlate with pCR, and current diagnostic modalities cannot distinguish with certainty between viable tumor and fibrosis. Because the decision to proceed with WW is based on cCR, it carries some uncertainty, since patients with initial cCR may later develop reappearance of clinically evident tumor.

WW is a strategy in patients with cCR following NAT, while organ preservation is the optimal outcome of this approach. Non operative management (NOM) is a term often used interchangeably with WW. We find NOM to be a confusing term, because operative local excision, which will not be discussed in this article, can be a part of a treatment algorithm for organ preservation.

Tumor regrowth is reappearance of neoplasia at the site of the treated tumor after a cCR. Regrowth is considered salvageable when curative treatment is possible. Non-salvageable regrowth or reappearance of cancer in the pelvis following a curative resection is considered a recurrence, but this term should not be used for a salvageable regrowth in cCR patients.

Adjuvant therapy is given after surgery, and neoadjuvant therapy is given before surgery. Neoadjuvant systemic chemotherapy can be added to CRT. Chemotherapy that is given before CRT is called induction chemotherapy, and chemotherapy given after CRT (and before surgery) is called consolidation chemotherapy. The full regimen of induction or consolidation chemotherapy (eight cycles of FOLFOX [leucovorin, fluorouracil, oxaliplatin] or five cycles of CapeOX [capecitabine, oxaliplatin]) along with CRT is called total neoadjuvant therapy.

History of WW

The ability of radiotherapy to cure rectal cancer without surgical intervention has been known since H. H. Janeway’s 1920 report of a 49-year-old woman whose rectal cancer was cured by contact radiation and subsequent implantation of radioactive seeds ¹⁰. For decades following this report, radiation was considered the primary curative treatment for rectal cancer and surgery was reserved for salvage following treatment failure ¹¹.

In modern times, A. Habr-Gama’s group pioneered WW of rectal cancer, as described in their report on 265 patients with resectable rectal cancer assessed for cCR at 8 weeks after CRT ¹² In that study, patients with incomplete clinical response underwent TME, and those with cCR entered a strict surveillance program that included in the first year monthly DRE, proctoscopy, measurement of the level of carcinoembryonic antigen in serum, and biopsy when possible. During the second and third year, surveillance was performed every 2 months and every 6 months, respectively. Seventy-one (27%) of the 265 patients had a cCR.

With mean follow-up of 57.3 months, three patients had distant metastasis and two patients had regrowth. Both patients with regrowth were successfully salvaged with local therapy. The 5-year rates of overall and disease-free survival were 100% and 92%, respectively. These results suggested that with a carefully structured surveillance program, deferral of surgery is safe, sustained cCR can be indicative of a cure, and regrowth can be effectively salvaged. Subsequent reports from specialized cancer centers around the world have provided additional evidence of the safety and efficacy of WW in LARC patients with cCR ^13–20. However, some practitioners are reluctant to adopt WW because of the lack of standardization in response assessment and the lack of data from a randomized control trial.

Patient Selection

Several studies on WW have reported outcomes specifically for LARC patients who received multimodal NAT as part of standard care ^13–15, ¹⁸. Patients with cCR to NAT are natural candidates for WW, given the often unsatisfactory quality of life in LARC patients after TME. The majority of LARC patients who have undergone NAT and TME have a constellation of symptoms including frequency, urgency, and fragmentation of bowel movements and incomplete evacuation that is commonly known as low anterior resection syndrome and which can persist for 15 years or longer ⁴. WW therefore offers the prospect of considerable potential benefit for LARC patients. The potential disadvantage of withholding a treatment modality from LARC patients is that the risk of disease progression is not negligible in this population.

Other studies have focused on WW in patients with early-stage (cT2N0M0) low rectal cancer ^{21, 22}. These tumors are not amenable to transanal resection alone, since local recurrence is found in as many as 42% of patients. Not uncommonly, these tumors are upstaged by pathological assessment to pT3, qualifying them as LARC ^{7, 23}. TME is the standard of care for patients with cT2N0M0 rectal cancer, but for a distal cT2N0M0 tumor, abdominoperineal resection (APR) with a permanent stoma is the only viable option. The prospect of avoiding a stoma by undergoing WW is highly desirable for patients. The potential disadvantages of WW for this population are the synergistic sequelae of radiotherapy and resection for patients with an incomplete clinical response (iCR) that could have otherwise been treated with resection alone.

Although patients receiving NAT for rectal cancer in any location are potential candidates for WW, patients that are likely to benefit the most are those with distal cancers requiring a low anastomosis or APR. Patients with large, circumferential, ulcerated tumours often develop a concentric scar that narrows the lumen of the rectum and prevents proper endoscopic evaluation of tumour response. These patients may not be ideal candidates for a WW strategy.

Regardless of the stage of the primary tumor, patients must adhere to a rigorous screening program over a period of many years and be able to manage the uncertainty regarding the status of their disease and treatment plan.

Treatment Regimens

Both short-course radiation therapy (SCRT) consisting of five fractions of 5 Gy (total, 25 Gy) and long-course CRT consisting of multiple fractions of 1.8 to 2 Gy (total, 40 to 50.4 Gy) combined with a sensitizing fluoropyrimidine have been shown to be effective in preventing local recurrence in patients with LARC ^{24, 1}. The interval to surgery after SCRT can be short (1 week) or extended (4 to 8 weeks), whereas the interval to surgery after CRT is usually 6 to 8 weeks after radiotherapy ^{1, 8}. In the majority of WW studies, participants were treated with CRT ^{12–17, 19, 20}. Likewise, a systematic review of WW studies evaluated only series in which the therapy was CRT ²⁵. In studies that assessed traditional CRT (40 to 50.4 Gy) without neoadjuvant systemic chemotherapy, the proportion of patients with cCR was between 11 and 61% ^{12, 15, 13, 8}.

Two prospective randomized control trials indirectly support the use of CRT in WW. The trials compared SCRT followed immediately by surgery with CRT followed by surgery 4 to 6 weeks after radiotherapy ^{26, 27}. These trials found no differences in the rate of local recurrence, distant metastasis, overall survival, or late toxicity, but the differences in pCR were striking (0.7% in SCRT patients vs. 16.1% in CRT patients in the Polish trial; 1% in SCRT patients vs. 15% in CRT patients in the Trans-Tasman trial). Even though no patients underwent WW in these trials, the findings imply that CRT is more appropriate for WW protocols. However, the differences in pCR rates between SCRT and CRT patients may be due in part to the shorter interval to surgery after SCRT and not solely due to possible differences in efficacy. While CRT is most commonly used in WW, there is limited data on the use of SCRT in WW as well. Two studies investigated SCRT and WW in a small group of elderly patients too frail to receive chemotherapy. In the first, a prospective study of 30 elderly frail patients, Bujko et al. ²⁸ reported that no patients experienced toxicity of grade III or higher. At a median of 10 weeks following radiotherapy, 20 (66%) patients had cCR, and all 20 entered WW. With a median follow-up of 21 months, one patient had a regrowth (no salvage information was reported). In the second study, Cummings et al. ²⁹ retrospectively analyzed data from 20 elderly patients too frail for CRT who received SCRT. Ten (50%) of the 20 patients underwent SCRT and surgery, and the other 10 (50%) patients received SCRT as definitive therapy in the context of WW. One patient experienced grade III toxicity. Of the 10 patients who underwent surgery, 3 (30%) had a pCR and 3 (30%) died in the early postoperative period. Of the 10 WW patients, none were treated with additional palliative procedures and 2 (20%) died from progression of disease. The surgery group and the WW group did not differ significantly in overall survival or disease-free survival. The authors of the two studies concluded that SCRT with WW is feasible in elderly frail patients.

One retrospective study investigated the use of SCRT with the addition of three to eight cycles of neoadjuvant FOLFOX (25 patients) or CapeOX (1 patient) ³⁰. No patients experienced toxicity of grade III or higher. Nine patients (35%) had a cCR, and six of them entered WW. With a median follow-up of 18 weeks, one patient experienced regrowth 20 weeks following initial restaging (no salvage information was reported). Of the 20 patients who underwent surgery, 7 (35%) had a pCR. Although these series were small and mostly retrospective, they do indicate that SCRT with or without consolidation chemotherapy can be a potentially valuable component in a WW protocol. Results of the RAPIDO trial (NCT01558921) ³¹, where patients are randomized to standard CRT followed by TME and optional adjuvant chemotherapy or to TNT comprised of SCRT and 6 cycles of CapeOX [capecitabine, oxaliplatin] consolidation followed by TME, may provide valuable information on the role of SCRT in WW.

Enhancing the Response to NAT

Response to NAT can be enhanced by increasing the radiation dose, giving systemic chemotherapy in the neoadjuvant rather than adjuvant setting in addition to radiotherapy, and/or delaying clinical assessment and surgery.

Radiation Dose

Appelt et al. ³² showed a highly significant dose-response relationship with radiation doses of 50.4 to 70 Gy in histological specimens from LARC patients. This dose range is higher than the range commonly used for LARC, and the authors concluded that it would be of interest for WW protocols. In a subsequent study of CRT in patients with T2-T3 N0-N1 disease considered for WW, Appelt et al. ¹⁶ delivered 60 Gy to the tumor, 50 Gy to lymph nodes, and an additional 5 Gy in endorectal brachytherapy. A remarkable 78% of patients had a cCR 6 weeks after the end of treatment. About a third of the patients had early-stage rectal cancer, which may have contributed to the high response rate.

Systemic Chemotherapy

In a retrospective cohort study, Cercek et al. ³³ compared the rates of pCR or sustained cCR for 12 months in 308 patients who received TNT with induction chemotherapy and 320 patients who received CRT. pCR or sustained cCR was seen in 37% of the patients who received TNT and only 21% of the patients who received CRT. Additionally, patients who received TNT were more likely to complete planned systemic chemotherapy with fewer dose reductions. In a prospective clinical trial, Garcia-Aguilar et al. ⁸ compared the rate of pCR between four sequential groups of patients who received CRT alone or in combination with two, four, or six cycles of FOLFOX. All systemic chemotherapy was administered after CRT, and no patients underwent WW. With each addition of two cycles of FOLFOX, the pCR rate increased: no FOLFOX, 18%; two cycles, 25%; four cycles, 30%; six cycles, 38%. The proportion of serious adverse events did not differ significantly between the four groups. The increase in the interval to surgery with the addition of FOLFOX cycles may have also contributed to the higher pCR rates.

Fokas et al. ³⁴ randomized 311 patients to either induction or consolidation systemic chemotherapy. All patients also underwent both CRT and TME. Consolidation therapy was associated with lower toxicity, higher compliance with CRT, and a higher pCR rate (25% vs. 17%; p < 0.001). These findings support the use of consolidation chemotherapy for WW. The anticipated results of the ongoing multi-institutional OPRA trial (NCT02008656), in which patients are randomly assigned to either induction or consolidation chemotherapy and then proceed to surgery or WW depending on response, may provide more conclusive evidence of the superiority of consolidation or induction chemotherapy.

Interval to Assessment

Tumor response to radiation is time dependent, and maximal tumor regression can take months. In a retrospective study of data from the National Cancer Database for more than 17,000 patients who underwent neoadjuvant CRT and resection, intervals of more than 8 weeks from CRT to surgery were associated with higher rates of pCR ³⁵. Delaying surgery increased the rate of pCR from around 0% at an interval of 1 week to around 11% at an interval of 10–11 weeks, after which the rates of pCR plateaued. Longer delays were also associated with lower rates of readmission within 30 days after surgery.

A systematic review and meta-analysis of 13 studies of nearly 20,000 patients found that intervals of >8 weeks from NAT to surgery were associated with higher rates of pCR than intervals of <8 weeks, with no significant difference in rates of complications ³⁶.

In the Stockholm III multicenter trial, patients who underwent SCRT with delayed surgery had lower yP stage, a higher rate of pCR, and higher tumor regression scores than patients who underwent surgery immediately after SCRT ^{37, 38}. The two groups did not differ significantly in recurrence-free or overall survival at 2 years after surgery. SCRT with delayed surgery was associated with fewer postoperative complications. In the study by Akgun et al.³⁹, in which patients were randomly assigned to undergo TME either within 8 weeks of CRT or >8 weeks after CRT, the rate of pCR was significantly higher in the latter group (18.6% vs. 10%; p = 0.03), with no significant difference in the quality of resection or in surgical morbidity.

Based on the available evidence, waiting at least 8 weeks following NAT to perform surgery is safe and efficacious and has the possible added benefit of decreasing postoperative morbidity.

Predictors of Response

Certain tumor- and patient-related factors have been shown to be associated with response to NAT. In a systematic review and meta-analysis of 18 studies totaling 1186 patients with pCR, Huang et al. ⁴⁰ showed that clinical factors associated with pCR include older age, smaller tumor size, short distance from the anal verge, and negative lymph nodes status. In a large retrospective study, Das et al. ⁴¹ found that the circumferential extent of the tumor, high level of carcinoembryonic antigen, and distance from the anal verge were inversely associated with tumor response.

Interest is growing in molecular and genetic predictors of response that would help tailor precise treatment plans for individual patients. Personalized treatment could help optimize oncological and functional outcomes in patients predicted to have a response while avoiding the morbidity and sequelae of NAT in patients not likely to have a response. Common genetic alterations that are associated with poor response to NAT include mutations in p53 and KRAS, which are found in about 70% and 40% of rectal tumors, respectively ^42–44. In contrast, microsatellite instability and mismatch repair deficiency, which are less common in rectal cancer (≈2% of patients) than in colon cancer, are associated with good response to traditional CRT⁴⁵.

The tumor immune microenvironment can also provide information on the likelihood of response to NAT. The presence of high levels of FoxP3+ regulatory T cells in the stroma of resected rectal tumor specimens was found to be associated with poor response ⁴⁶. The presence of CD8+ tumor-infiltrating lymphocytes both in pretreatment biopsies and in resected specimens was associated with good response, whereas their presence in the stroma surrounding the tumor in pretreatment biopsies was not associated with response ^{47, 48}.

Although the search for new molecular predictors of response to NAT has been very active and holds great promise, no clinical, radiological, or molecular test is currently available for accurately predicting pCR from cCR.

Assessment of Response

Identifying true responders is both difficult and critical for a successful WW protocol. Selection of patients suitable for WW remains a challenge. The goal is to correctly identify patients whose surgical specimen would show pCR were they to undergo TME. The relationship between cCR and pCR is imperfect and can be fraught with problems related to the sensitivity and specificity of cCR assessment. In some WW series, 8 to 15% of patients with incomplete clinical response ended up having a pCR, indicating imprecise specificity ^{12, 49}. Conversely, in the first 2 years of follow-up surveillance, regrowth occurs in 15 to 25% of patients who enter WW, indicating imprecise sensitivity ^{50, 51}. If the criteria of cCR are too strict, some patients with complete response may end up undergoing surgery needlessly. If the criteria are too loose, some cases of residual tumor may be missed.

One of the challenges for broad implementation of WW is establishing uniform and reproducible criteria for tumor response. Three modalities can be used to assess response: DRE, MRI, and endoscopy. Each modality is accurate but imperfect. Combining modalities increases accuracy. In their investigation of the accuracies of DRE, endoscopy, and MRI in predicting pCR or sustained cCR, Maas et al. ⁴⁹ found that clinical assessment was the most accurate. When all three modalities were consistent with absence of residual tumor, the accuracy of predicting complete response was 98%.

Patients with a very significant response that does not meet all criteria of a cCR, called a near complete clinical response (ncCR), can be entered in an intensive surveillance protocol, as long as the tumor continues showing signs of ongoing response until all strict criteria of a cCR are achieved. While not the norm, it may take up to a year for some tumors to achieve a cCR. For this reason, a three-tiered response assessment schema is currently being tested in the OPRA trial and consists of DRE, endoscopy, and T2- and diffusion-weighted MRI. Based on that assessment, patients are considered having a cCR, an iCR, or a ncCR. Descriptions of the clinical and imaging findings in cCR, iCR and ncCR can be seen in Table 1. Studies aimed at validating the reproducibility of that response assessment schema are underway. Endoscopic examples of cCR, iCR and ncCR can be seen in Figure 1.

Table 1.

Response Assessment Schema in the OPRA Trial

	Complete Response	Near Complete Response	Incomplete Response
Endoscopy	Flat, white scar Telangiectasia No ulcer No nodularity	Irregular mucosa Small mucosal nodules or minor mucosal abnormality Superficial ulceration Mild persisting erythema of the scar	Visible tumor
Digital Rectal Exam	Normal	Smooth induration or minor mucosal abnormalities	Palpable tumor nodules
MRI-T2W	Only dark T2 signal, no intermediate T2 signal AND No visible lymph nodes	Mostly dark T2 signal, some remaining intermediate signal AND/OR Partial regression of lymph nodes	More intermediate than dark T2 signal, no T2 scar AND/OR No regression of lymph nodes
MRI-DW	No visible tumor on B800-B1000 signal AND/OR Lack of or low signal on ADC map Uniform, linear signal in wall above tumor is ok	Significant regression of signal on B800-B1000 AND/OR Minimal or low residual signal on ADC map	Insignificant regression of signal on B800-B1000 AND/OR Obvious low signal on ADC map

T2W, T2 Weighted; DW, Diffusion Weighted

Figure 1.

Legend: Endoscopic Examples of Complete (A) Near Complete (B) and Incomplete (C) Clinical Response

Clinical and radiological modalities underestimate response to NAT. Guillem et al. ⁵² found that DRE underestimates response in 78% of patients, relative to response in pathological specimens. Smith et al. ⁵³ reported that 74% of patients with pCR had a residual mucosal abnormality that would have led to their being categorized as having had an incomplete clinical response. Maas et al. ⁴⁹ found that of the patients for whom DRE, endoscopy, and MRI all indicated the presence of residual tumor, 15% had pCR or sustained cCR. Using a pattern-based MRI approach to predict complete response, Lambregts et al. ⁵⁴ were able to achieve a sensitivity of 94% and a specificity of 77%.

In current WW practice, sensitivity appears to supersede specificity: we err on the side of having too many patients with complete response undergoing TME. This approach seems sensible, since TME is the accepted standard of care worldwide. To counterbalance imperfect sensitivity, strict and frequent follow-up is needed to allow residual tumor to be diagnosed in a timely manner, when it is still salvageable by TME.

Long-Term Monitoring of Response

Surveillance protocols for patients entering WW vary but usually include clinical examination, measurement of the level of carcinoembryonic antigen in serum, endoscopy and MRI for assessment of local recurrence, and CT for assessment of distant recurrences. The vast majority of cases of regrowth occur in the first 2 years after treatment, and monitoring should therefore be intensive during this period. In years 3-5 following treatment the surveillance should be less intensive, and after year 5 patients should enter survivorship surveillance.

The surveillance protocol for the OPRA trial includes DRE, level of carcinoembryonic antigen, and endoscopy every 4 months for the first 2 years following treatment and then every 6 months. MRI is performed every 6 months for the first 2 years after treatment and yearly thereafter. CT of the chest, abdomen, and pelvis is performed yearly for 5 years (Table 2). This demanding follow-up makes WW reasonable only for patients that are compliant and that have easy access to appropriate healthcare providers.

Table 2.

Long Term Surveillance in the OPRA Trial

Months after Treatment	3-6	9-12	15-18	21-24	30*	36*	42*	48*	54*	60*
History and Physical	√	√	√	√	√	√	√	√	√	√
Endoscopy	√	√	√	√	√	√	√	√	√	√
MRI Rectum	√	√		√		√		√		√
CT CAP		√		√		√		√		√
CEA	√	√	√	√	√	√	√	√	√	√

CAP, chest abdomen and pelvis; CEA, Carcinoembryonic Antigen

^*+/− 30 days

Long-Term Outcomes

Oncological Outcomes

Currently, there is no prospective data on WW. All data on oncological outcomes is based on retrospective series with variable treatment regimens and indications and without uniform criteria of staging, patient selection, and of who enters WW. The timing of assessment is also inconsistent. Often the denominator is unclear because studies start with a cohort of patients that have already achieved cCR and not with all those entering NAT. It is also unlikely that prospective randomized data will be available, since it will be difficult for patients to agree to be randomized into surgery versus organ preservation.

In their systematic review and meta-analysis of 23 studies totaling 867 patients who underwent WW, Dossa et al. ⁵¹ determined that the rate of regrowth at 2 years was 15.7% and the rate of salvage surgery after regrowth was 95%. The authors compared nonregrowth recurrence, cancer-specific mortality, disease-free survival, and overall survival between patients who underwent WW after cCR and patients with pCR following TME. These outcomes did not differ significantly between the two groups, except for disease-free survival, which was longer in patients with pCR following TME. However, non-regrowth disease free survival and cancer specific mortality was not different between the groups. The authors estimated that of 1000 patients who undergo WW, 2 would not be candidates for salvage surgery because of locoregional disease and 1 would not be a candidate because of distant metastasis. In a systematic review by Kong et al. ⁵⁵ of nine studies totaling 370 patients who underwent WW, the rate of regrowth was 28%, and 84% of those regrowth cases were salvaged with surgery. The rate of distant metastasis in WW patients did not differ significantly from the rate in patients with pCR following TME (5.9% vs. 4.3%, respectively; p = 0.59). The two groups of patients also did not differ significantly in disease-free survival or overall survival.

In 2014, the International Watch & Wait Database was created in order to pool existing retrospective data and prospectively accumulating data into a centralized resource. Using this database, van der Valk et al. ⁵⁰ reported in 2018 their analysis of 880 patients who underwent WW at 47 treatment centers across the globe. Their large heterogeneous dataset provided a reliable picture of the real-world risks and benefits of WW. The 2-year incidence of regrowth was 25%; 88% of all cases of regrowth were diagnosed in the first 2 years following treatment, and 77% of the regrowth cases were salvaged by TME. Distant metastasis was diagnosed in 8% of patients. Overall survival and disease-specific survival were 85% and 94%, respectively.

Certain subpopulations of patients who undergo WW may have a higher risk of progression. Smith et al. ²⁰ reported a significantly higher rate of distant metastasis in patients who had regrowth during WW than in patients who did not have regrowth (36% vs. 1%; p <0.001). The authors noted that the higher rate may be due either to deferral of surgery or to the fact that regrowth may be associated with aggressive tumors that carry a higher risk of metastatic progression.

Organ Preservation and Functional Outcomes

The rates of rectum preservation in patients who undergo WW range between 70 and 82% ^{19, 16, 15, 20}. Applet et al. ¹⁶ reported excellent functional outcomes in WW patients 2 years after treatment, with 69% of patients reporting no fecal incontinence (median Jorge-Wexner score, 0). Martens et al. ¹⁵ reported major incontinence in 14% of patients at 3 years after treatment. A subset of patients with incomplete clinical response at initial response assessment underwent local excision as an adjunct to WW. These patients had worse functional outcomes, with 43% experiencing major incontinence, in contrast to 4.5% of patients who did not need local excision.

Patient and Surgeon Perspectives on WW

In a 2019 study by Gani et al. ⁵⁶, in which questionnaires were completed by 49 patients diagnosed with rectal cancer who were awaiting multimodal treatment, 83% of the patients said that they would consider WW in case of cCR. The vast majority were willing to accept a 2-year regrowth rate of 25% and an intensive follow-up protocol (94 and 96% of patients, respectively). Equivalent cure rates were a prerequisite for entering WW for 55% of patients, but 30% of patients were willing to accept a long-term cure rate of 88% instead of 90%, and 11% of patients still preferred a nonoperative approach if it was associated with a cure rate of 80% instead of 90%. It appears that outcome measures such as overall survival and disease-free survival leave out some important biopsychosocial consequences of cancer diagnosis and treatment.

In a 2018 study by Schwartzberg et al. ⁵⁷, in which questionnaires were completed by 452 colorectal surgeons from the United States, Europe, Australia, and Brazil, adoption and acceptance of WW among the surgeons was high. However, adoption and acceptance of WW among members of the American Society of Colon and Rectal Surgeons was significantly lower than among members of societies of colorectal surgeons in the other three regions (41% vs. 75%; p < 0.001).

Conclusion

Rectal cancer patients with a complete response to NAT may not benefit from TME, but identifying true responders is challenging. Evidence supporting WW is based on retrospective case series, and prospective data are needed. Since WW is acceptable to most rectal cancer patients and already being demanded by some, it should be part of the treatment discussion. To enter WW, patients must be well-informed and willing to undergo an intensive surveillance protocol. A prospective trial with a rigorous protocol and objective assessment standards may currently represent the optimal setting for patients to undergo WW.