In 1982, Bill Heald proposed the total mesorectal excision (TME) surgery, reducing the local recurrence rate from 30–40% to 3.7% [1]. To this day, TME remains the cornerstone of rectal cancer radical therapeutics. Although TME demonstrates a remarkable ability to control local tumors, the postoperative complication rate (50.6%) always has troubled surgeons and patients [2]. Especially for patients with low rectal cancer, TME always parallels a lower organ preservation rate and quality of living reduction. Since 2004, the projects NSABP R01-03 [3,4,5] studied by the American National Surgical Adjuvant Breast and Bowel Project (NASBP), and a series of research CAO/ARO/AI [6] made in Germany established the status of neoadjuvant chemoradiotherapy (nCRT) as the standard treatment for rectal cancer. Then Habr-Gama et al. [7] proposed the strategy that among the T2-4/N0-2/M0 distal rectal cancer, the patients reached complete clinical response (cCR) standard after nCRT had the opportunity to be free from surgery as the prognosis was statistically the same in cCR patients with or without surgery. Subsequently, the strategy was named as “watch and wait” strategy (W&W).

According to current clinical studies, W&W is widely recognized for its higher cost-effectiveness [8], organ preservation rate, and quality of living compared with upfront TME. However, this strategy needs to be rationally taken as it proves to be worse in functional organ change, recurrence rate, and metastasis rate [9]. The process has been coming up for over 20 years, but it is still under global debate with many undefined details. The arguments mainly focus on the following aspects: (1) the W&W long-term oncological advantage compared with TME; (2) the method of advanced predicting the rectal tumor regression grade before nCRT to reduce unnecessary overtreatment; (3) the method to promote the nCRT efficacy and expand the number of cCR or pCR; (4) the surveillance strategy which is the most beneficial and convenient for the patients. This article discusses the above four details and explores the opportunities and challenges in W&W.

1 Comparison of "watch and wait" and "TME" oncologic outcomes

In clinical practice, clinicians are more interested in whether the oncologic outcomes of patients with rectal cancer receiving W&W are better than those of TME. In 2004, Habr-Gama et al. [10] were the foremost to publish a retrospective cohort study of patients with rectal cancer who achieved cCR/pCR after nCRT performing W&W compared with radical surgery. Of the 22 patients who underwent radical surgery were found to have achieved pCR, three patients incurred metastases, none developed recurrence, and the 5-year overall survival (OS) and disease-free survival (DFS) rates were 88% and 83%, respectively. Meanwhile, the 5-year OS and DFS rates of the 71 patients in the watch-and-wait group were 100% and 92.0%, respectively. At a median follow-up of 57.3 months, three patients suffered metastases and two patients developed recurrences, and the patients with recurrences did not develop recurrences or metastases at the 5-year follow-up after salvage surgery or radiotherapy, demonstrating the promising potential of the W&W strategy as an alternative to TME in patients achieved CR.

We analyzed 10 studies comparing the prognosis of patients who achieved cCR/pCR after neoadjuvant therapy underwent W&W versus TME (Table 1) [10,11,12,13,14,15,16,17,18,19] and observed that the OS rate of patients in the W&W group ranged from 71.6% to 100%, while the TME group had an OS rate of 88.0%-100%. We noticed that DFS rates ranged from 27.8% to 92.0% in the W&W group and from 82.8% to 98.0% in the TME group, respectively. This large difference was probably due to selection bias, the insufficient number of patients included, and the duration of follow-up. The rates of distant metastasis and recurrence were 0–15.5% and 2.8–26.7%, respectively, in the W&W group, and 3.7–13.6% and 0–3.9%, respectively, in the TME group. TME exhibited an excellent ability to control local recurrence of the tumor. Yu et al. [20] conducted a meta-analysis including 9 studies with 339 patients who received W&W and 479 patients who underwent TME and found no statistical difference in distant metastasis rates, 2-, 3-, and 5-year OS and DFS between the W&W and TME groups. There was no statistical difference between the W&W group and the TME group in terms of distant metastasis rate, 2-year, 3-year, 5-year OS and DFS, but the local recurrence rate was higher in the W&W group than in the TME group (OR = 8.54, 95% CI 3.52–20.71, P < 0.001). Nonetheless, Sander et al. [21] reported that 97% of patients with local recurrence after W&W were able to receive salvage radical surgery, and the 2-year local recurrence-free rate was 97.8% in patients who underwent salvage surgery.

Table 1 W&W versus TME for patients who achieved cCR/pCR after neoadjuvant therapy
Full size table

In addition, some investigators have proposed local excision for patients with cCR as an intermediate treatment option to balance W&W and TME [22]. However, we have to consider the potential issue of local resection. For one thing, there was no statistical difference in the rates of local recurrence (OR = 1.24, 95% CI 0.33–4.69, P = 0.75) and distant metastases (OR = 1.46, 95% CI 0.49–4.35, P = 0.50) between patients who underwent W&W and local excision after cCR [23]. For another, organ function and quality of life were worse in patients with local resection compared to W&W. Moreover, the quality of salvage TME surgery was worse in patients who had undergone local excision when they suffered local recurrence [24]. Although local excision has been proposed to predict lymph node metastasis at mesorectum, it has been known to have several additional issues. One is the wound healing problem. Irradiated tissue has a tendency to delay healing, possible infection, and more complications(Perineal Wound Complications After Extralevator Abdominoperineal Excision for Low Rectal Cancer). Another issue will be the difficulty of differentiating between fibrotic scar and local recurrence during surveillance.

A questionnaire survey of patients with rectal cancer revealed that their three main concerns were survival time, whether they would suffer from recurrence, and whether they would have an enterostomy [25]. In the absence of significant differences in OS and DFS, the W&W strategy, which preserves rectal function, has a considerable advantage over TME. It should be kept in mind that the current studies compared the prognosis of patients who achieved cCR (approximately 20%) with the W&W strategy and TME, whereas those who underwent nCRT and still required TME had worse organ function than those who underwent TME alone [26].

2 Strategies to improve the rate of cCR after nCRT

2.1 Extending the interval to surgery

At present, the optimal time for cCR evaluation in W&W strategy is still controversial. Studies have suggested that delayed clinical responses may occur in some cases, and that by extending the interval time to surgery, more patients may achieve cCR. In 2019, Habr-Gama et al. [27] retrospectively analyzed 49 patients with cCR after nCRT and found that it took 18.7 weeks on average to achieve cCR, and only 38% of patients achieved cCR within 10 to 16 weeks after nCRT. In addition, the time required for cT3b-d/cT4 patients was significantly longer than that for cT2/cT3a patients. However, extending the interval may also increase concerns about local tumor progression and distant metastasis, as well as the difficulty of remedial surgery. The latest NCCN guidelines recommended an interval of 6–11 weeks [28]. Chinese expert consensus suggested that 8–12 weeks after the end of neoadjuvant therapy should be used as the evaluation time of cCR, and for those with strong anal preservation intention and receiving intensive nCRT, it should be extended to 16–24 weeks as appropriate [29]. However, this time may not be the standard time point for determining cCR. The optimal interval for observation should be comprehensively considered in combination with patients' wishes, disease control risks and the difficulty of remedial treatment.

2.2 Total neoadjuvant therapy

In order to further improve the efficacy of patients with locally advanced rectal cancer (LARC), multiple clinical studies have advanced postoperative adjuvant chemotherapy to pre-surgery, namely total neoadjuvant therapy (TNT) [30]. The randomized clinical trial (RCT) study conducted by Conroy et al. [31] also showed that TNT could improve DFS and pCR rates in patients with LARC (HR = 0.69, 95%CI: 0.49–0.97, P = 0.034; 28% vs. 12%, P < 0.01). In the STELLAR [32] study from China, of 465 patients who underwent surgery, 91.5% in the TNT group and 87.8% in the CRT group underwent R0 resection (P = 0.19). In addition, ypN0 was present in 71.1% of the TNT group and 68.7% of the CRT group (P = 0.58). The total pCR and cCR rate of TNT group were 21.8%, which was significantly higher than that of CRT group (12.3%, P = 0.002). A meta-analysis involving 2,430 patients in 9 studies showed that the pCR rate was increased in TNT compared to nCRT in LARC patients [33]. Based on the above study results, it indicated that TNT could improve the surgical efficacy, reduce the tumor load and bring more survival benefits by enhancing the intensity of preoperative treatment for patients with LARC, but more high-quality RCT further verification is still needed.

2.3 Immune checkpoint inhibitor

Immune checkpoint inhibitor (ICI) is a promising breakthrough for precision treatment in rectal cancer. The degree of immune response is a predominant prognosis factor, and no evidence of relapse was noticed during the W&W follow-up period in patients with high immune score in histopathological biopsies [34]. Besides, among particular genotypes, especially microsatellite instability (MSI)/mismatch repair deficient (dMMR), the efficacy of ICI showed superiority over chemoradiotherapy in rectal cancer adjuvant therapy. André T. et al. managed two metastatic colorectal cancer (mCRC) with MSI-H groups administrated by PD-1 inhibitors and standard chemoradiotherapy, respectively. The progression-free survival (PFS) significantly increased in the PD-1 group (16.5 months vs. 8.2 months) [35], and the early administrating time of PD-1 theoretically correlated with an excellent prognosis. In recent years, the number of small-sample W&W clinical research on dMMR/MSI-H with ICI monotherapy is gradually increasing. After PD-1 inhibitor monotherapy, two dMMR patients with LARC [36] in 2019 and 12 dMMR patients with rectal adenocarcinoma [37] in 2022 were all determined to be cCR or pCR, and there was no evidence of relapse and metastasis. Worthy notice, achieving cCR in all 12 patients with dostarlimab monotherapy proves that the surgery might be overtreatment for a specific population, which is the same philosophy as precision therapy.

3 Tumor response prediction of neoadjuvant therapy

Although preoperative chemoradiotherapy will benefit many patients, especially those who achieve cCR, it remains unclear how to define the beneficiary population previously to reduce the over-nCRT medication to non-responders. Recent research reported a handful of biomarkers to predict sensitivity to nCRT for lower rectal cancer, including the single molecular markers HER-2, PI3K, and MSI molecular phenotypes [38,39,40]. In addition, tumor regression grades (TRG) are considered a meliorating prognostic indicator for patients with LARC receiving nCRT [41, 42]. Except for the molecular biomarkers like HIF-1α, p21, and CD133 [43], the imaging analysis can also predict the CR. One study reported that the MRI radiomics model can predict pCR after chemoradiotherapy [44], and another study reported that early FDG-PETCT assessment could predict the pathologic response and TRG of preoperative treatment in LARC. It may guide the revision of treatment strategy during preoperative nCRT [45].

Besides, the non-invasive liquid biopsy tests fit the practical needs of W&W forecast and follow-up. The degree of circulating tumor DNA (ctDNA) haplotype methylation in colorectal cancer significantly differed from the average population [46]. And 3-year recurrence-free survival rate in postoperative ctDNA-positive patients with adjuvant therapy was dramatically lower than in ctDNA-negative patients without adjuvant therapy (86.4% vs. 92.5%) [47], suggesting a close link between ctDNA and colorectal tumorigenesis and recurrence. Moreover, Wang et al. combine ctDNA or circulating cell-free DNA (cfDNA) baseline level with MRI tumor regression grade (mrTRG) to predict the response level precisely.

The first model to predict the odds of patients achieving pCR after nCRT consisted of four factors: the ctDNA baseline level, ctDNA clearance rate, ctDNA mutation rate, and mrTRG. The model’s area under the curve (AUC) was 0.886 in the internal validation [48]. Then another pCR prediction algorithm model was constructed by cfDNA combined with mrTRG, and the AUC of the training and test set were 0.92 and 0.92, respectively [49]. However, such studies generally have some limitations, including relatively short follow-up periods and the absence of independent external validation. Further comprehensive studies are needed to improve the prediction in clinical usage.

4 Strategies for follow-up and surveillance

It is generally believed that follow-up strategies should be formulated based on the pattern characteristics of local recurrence and distant metastasis after treatment for patients treated with W&W strategy. Many studies have confirmed that the most common treatment failure mode for the W&W population is local tumor regrowth, while the risk of distant metastasis is low [17]. The local regrowth of the tumor is mainly indicated by the regrowth of tumors at sites of the primary intestinal lumen, and a few patients show new enlarged lymph nodes in the mesorectal region [50]. About 88% of local regrowth was diagnosed two years after the beginning of W&W strategy [51], which suggests that the first 2 years after receiving W&W strategy is a high-risk period for local regrowth, and intensive monitoring is required. Habr-Gama and colleagues [50] advocated that the follow-up examination program and frequency of W&W strategy include digital rectal examination, colonoscopy, and CEA level measurement (once every 2 months in the first 2 years; once every 3 months in the third year; once every 6 months after 3 years); pelvic imaging examination (CT, MRI or endorectal ultrasound) and CT scan of the chest and abdomen (once every 6 months in the first 2 years; once every year after 2 years). Moreover, the follow-up strategy adopted by the W&W group in the British OnCoRe project included digital rectal examination and MRI (once every 4–6 months in the first 2 years; the follow-up frequency after 2 years is the same as that of the radical resection group), as well as colonoscopy, CT scan of the chest, abdomen and pelvis, CEA detection (the follow-up frequency is the same as that of the radical resection group) [52]. Chinese expert consensus recommends follow-up every three months for the first three years [29] (Table 2).

Table 2 Strategies for follow-up and surveillance of W&W
Full size table

A retrospective multicenter registry study conducted by the International Watch & Wait Database (IWWD) enrolling 793 patients who underwent W&W strategy reported that rectal cancer patients who achieved sustained cCR for 3 years after neoadjuvant therapy had a ≤ 5% risk of local regrowth [53]. This finding suggests that intensive surveillance of the rectum to detect local regrowth may no longer be necessary after maintaining cCR for more than 3 years and further follow-up can be performed according to the follow-up plan for patients treated with radical resection.

5 Salvage strategy of tumor regrowth

5.1 Surgical salvage treatment

Salvage TME surgery is the preferred treatment when local tumor regrowth occurs in patients receiving W&W strategy. A study in the IWWD database showed that among 148 patients with tumor regrowth after receiving W&W strategy, 115 (78%) underwent salvage TME surgery, of which 101 (88%) had tumor negative resection margins (R0 resection), that is, the salvage rate reached 88% [51]. In addition, when the locally regenerated tumor is small and superficial and there is no sign of involved lymph nodes on MRI, local tumor resection can also be considered an alternative to salvage TME surgery.

So far, three studies have reported the salvage effect of local excision on local tumor regrowth. Sande et al. [54] retrospectively analyzed the type and outcome of salvage treatment for patients with local regrowth after receiving W&W strategy. The results showed that 84 (94%) of 89 patients with local regrowth received salvage surgery, of which 58 (69%) underwent TME surgery, and 26 (31%) underwent local excision. Among 26 patients who underwent local excision, 5 patients underwent TME surgery due to poor histology later. After a median follow-up of 20 months, 2 patients (9.5%) had local recurrence after local excision alone, while 2 patients (3.4%) had local recurrence after TME surgery [10]. Laura et al. [55] analyzed the survival outcome of 73 patients with local regrowth and found that there was no statistical difference in the local recurrence rate between patients with local regrowth who received salvage TME surgery (n = 43) and salvage local excision (n = 22) [11]. Geubels et al. [56] evaluated the survival results of 77 rectal cancer patients in a large W&W cohort who received local excision for salvage treatment due to suspected local regrowth of which 13 patients (ypT2-3 and/or non-radical resection) received TME surgery after local excision. During a median follow-up of 53 months, 14/64 (21.9%) patients who were treated with local excision alone had local recurrence, and they were successfully treated with salvage TME surgery without recurrence. In general, the 5-year total organ preservation rate of the 77 patients receiving local excision salvage treatment was 63%, the colostomy-free survival rate was 68%, and the total survival rate was 96%. These studies suggest that local excision may be a safe alternative to salvage TME surgery for patients with local regrowth under specific circumstances.

5.2 Non-operative salvage treatment

Brachytherapy is a potential option for locally regenerated patients who strongly refuse to undergo salvage treatment through surgery. As reported, patients with T1N0 rectal cancer can obtain high local control through contact X-ray brachytherapy, and there is no serious toxic effect [57,58,59]. It should be noted that when salvage brachytherapy is performed on patients with local regrowth, the dose of radiotherapy received by patients in the past must be considered [60].

6 Conclusion

In conclusion, the W&W strategy has obvious advantages, which can improve the efficacy of comprehensive treatment for LARC, significantly increase the rate of anus preservation, avoid surgical complications, and ensure the quality of life of patients. If the patients develop tumor regrowth during W&W, most can still be saved by definite surgical resection. However, the cCR rate after nCRT is still low, and radiotherapy also has its drawbacks, leading to partial functional limitation of the anus in some patients. The current issues, including improving cCR/pCR rates, finding efficient predictors, standard follow-up planning, and the rescue surgery method, still require larger sample size and more standardized clinical trials to produce credible evidence. Therefore, we must look rationally at the cCR after nCRT for low and middle rectal cancer, understand the risks of the W&W strategy, and make a reasonable choice. In particular, we should actively carry out prospective clinical trials to produce high-level evidence suitable for Chinese characteristics so that more rectal cancer patients can benefit from W&W.