Achieving a Cure Without Total Mesorectal Excision in Rectal Adenocarcinoma

Abstract

The Oncology Grand Rounds series is designed to place original reports published in the Journal into clinical context. A case presentation is followed by a description of diagnostic and management challenges, a review of the relevant literature, and a summary of the authors’ suggested management approaches. The goal of this series is to help readers better understand how to apply the results of key studies, including those published in Journal of Clinical Oncology, to patients seen in their own clinical practice.

Rectal cancer is a curable disease, yet curing the disease can be associated with lifelong morbidity because of the nature of the curative-intent treatment strategies. A major focus of modern prospective trials has been to maintain current cure rates, while minimizing lifelong lifestyle alterations and maximizing quality of life. Navigating the complex landscape of therapeutic options for rectal adenocarcinoma with a focus to accomplish this quality-of-life improvement is a critical focus area for future clinical trials. Many challenges remain on the path to optimizing cure and minimizing morbidity, and include improving initial staging accuracy, more precise selection of neoadjuvant therapy used for each patient, choosing the optimal surgical management strategy, and ensuring modern radiation therapy approaches are being used. Finally, organ preservation strategies have moved to the forefront in the management of both early and locally advanced rectal cancers and hold the potential for significant changes to come for patients with rectal cancer. Herein, we highlight some of the challenges remaining in the field, progress made, and how the recent data from the Canadian Cancer Trials Group phase II trial can be put into context with the ACOSOG Z6041, CARTS, and GRECCAR2 trials.

CASE PRESENTATION

A 47-year-old woman presents with a newly diagnosed rectal tumor found on her first screening colonoscopy. Biopsy was obtained showing invasive adenocarcinoma arising in association with tubulovillous adenoma and high-grade dysplasia. Mismatch repair proteins were retained, and carcinoembryonic antigen was 1.23. Computed tomography of the chest, abdomen, and pelvis showed no evidence of metastatic disease. Rectal magnetic resonance imaging (MRI) was notable for rectal motion artifact and presence of stool and gas, limiting evaluation. The tumor was noted from the 12 to 6 o'clock position, and the craniocaudal extent was 2.6 cm; it was semiannular in shape and did not reveal any mucinous features. The distance of the lowest extent of the tumor from the anal verge was 5.1 cm and from the anorectal junction was 2.0 cm. The tumor was below the anterior peritoneal reflection and immediately adjacent to the posterior vaginal wall with no visible intervening fat plane. Staging was described as an American Joint Committee on Cancer eighth edition T1 tumor invading the submucosa. The mesorectal fascia was abutted anteriorly; however, there were no tumor spiculations close to the mesorectal fascia. Lymph nodes were examined and two mildly prominent perirectal lymph nodes were near the tumor. Lymph node 1 was seen at the level of the tumor with smooth border and homogeneous signal intensity, and its size was 3 mm. Lymph node 2 was located above the level of the tumor, at the 2 o'clock position, with smooth borders, and homogeneous signal intensity, and the size was again 3 mm Figures 1A and 1B highlight findings from the MRI at the time of diagnosis. Figure 1C shows a left, perirectal node thought to be benign. The patient was presented at multidisciplinary tumor board.

FIG 1.

Primary tumor 3 Tesla MRI: (A) sagittal and (B) short axis T2-weighted MR images. The primary tumor is difficult to identify in the mid-rectum on sagittal images (red arrowheads). On short axis, there is fullness in the left anterolateral mid-rectum (red arrowheads), either the primary tumor and/or mucosal fold. An adjacent tiny focal left perirectal spiculation (white arrow) was felt to likely be a vessel rather than tumor. (C) Perirectal node: short axis T2-weighted MR demonstrated a 3-mm left perirectal lymph node (white arrow) that was thought to be benign. This node along with the subtle irregular primary tumor was staged as AJCC stage T1N0. Images courtesy of Parag P. Tolat, MD, Medical College of Wisconsin. AJCC, American Joint Committee on Cancer; MRI, magnetic resonance imaging; N, node; T, tumor stage.

CHALLENGES IN DIAGNOSIS AND MANAGEMENT

Rectal adenocarcinoma is a curable disease, yet such cures can be associated with lifelong morbidity because of the nature of the curative-intent treatment strategies. A major focus of modern prospective trials is to maintain current cure rates, while minimizing lifelong lifestyle alterations and maximizing quality of life (QoL). Navigating the complex landscape of therapeutic options for rectal adenocarcinoma with a focus to accomplish this QoL improvement is a critical focus area for clinical trials.

The initial staging of patients with rectal cancer presents the first opportunity, and challenge, to select a specific pathway of optimal curative intervention. The current standard of care includes proceeding directly to total mesorectal excision (TME), with either low anterior resection (LAR) or abdominoperitoneal resection (APR), for patients with T1/2N0M0 rectal adenocarcinoma.¹ For patients who are deemed to require an APR, neoadjuvant therapy can be recommended regardless of stage, with the goal of downstaging the primary tumor. Such downstaging can result in organ preservation (OP) if the tumor responds completely, or a potentially less extensive surgical resection.² Precisely how this neoadjuvant therapy is given is rapidly evolving, and the subject of extensive ongoing research, yet relies on staging accuracy as the initial branch point. The current standards of staging were historically on the basis of endorectal ultrasound, however have largely been supplanted by MRI.^1,3,4

A first important challenge facing rectal adenocarcinoma management are the limitations of current staging methods. Indeed, when examining both the sensitivity and specificity of rectal cancer staging using modern MRI, there are limitations to this approach that must be recognized.^5,6 Such limitations introduce the potential for overtreatment if a total neoadjuvant therapy or TNT approach is selected for a tumor that is thought to be more advanced than it really is (eg, all treatment given before radical surgery for a cT3 tumor when in fact it was a cT1 or cT2 tumor). This is particularly notable in patients with early-stage disease. For example, in a Dutch population-based study, a considerable number of tumors were felt to be early stage (cT1 to T2N0), and therefore potentially suitable for local excision (LE). A large number of patients were overstaged, with half having an incorrect T stage.⁷ Interpretation of rectal MRI requires considerable skill and experience, which introduces potential for disparities in care depending on center volume and experience.⁸ The potential disparity raises issues regarding patient selection for neoadjuvant therapy using either chemotherapy or radiation (RT), with several series questioning the correlation between clinical and pathologic staging.⁹ Initial staging is thus critical to patients for the optimal selection of the most appropriate and least life-altering modality that will cure their malignancy. Although these challenges are well understood, robust and widely available solutions to overcome them remain elusive. There are encouraging signals that novel strategies to more accurately stage patients with rectal adenocarcinoma using radiomics may be emerging; however, these approaches are not ready for prime time.¹⁰ A critical and unmet need in the management of rectal adenocarcinoma is improvement in pretreatment staging, which is essential for selecting the optimal neoadjuvant therapeutic approach. The multidisciplinary oncologic team must recognize these limitations, and the important role of obtaining the most precise staging possible before proceeding with definitive management.

A second important challenge is precisely what type of neoadjuvant therapy should be used? Historically speaking, patients have been treated successfully with neoadjuvant concurrent chemotherapy and RT, followed by evaluation for surgical resection.¹¹ Such RT can often be given as a short course, with only five treatments over a single week.¹² Although this is an appealing strategy in one regard, it also has limitations in that some patients are unable to tolerate systemic chemotherapy in the postoperative setting. Approaches to rectal cancer have now accounted for this, with patients often being treated with TNT, ushering chemotherapy into the neoadjuvant setting before consideration for the final surgical resection.¹³ Although a TNT approach can increase the completion of all therapy, and improve complete response, as noted above, it can also introduce the potential for over treatment.¹⁴

The third challenge is selecting the optimal surgical management strategy. This can range from no surgery at all, to LE, to TME approaches including an APR or an ultra-low LAR with coloanal anastomosis. Avoiding unnecessary surgery is important as the associated risks and morbidity are not trivial. However, it must be recognized that surgery is both a therapeutic and diagnostic modality for rectal adenocarcinoma, and this should not be ignored. Precision pathologic diagnosis has played an important role in rectal cancer management and risk stratification. Moreover, in the past several years, surgery is continuing to evolve for patients with rectal cancer. Novel surgical techniques are emerging that may reduce morbidity. Such techniques include robotic surgical technology, transanal TME, and robotic/laparoscopic lateral pelvic lymph node dissections.¹⁵

A final challenge is the precise RT modality that should be applied. The delivery of RT is rapidly evolving. Historic RT delivery techniques have resulted in poor bowel control, urinary incontinence, and sexual dysfunction.^16,17 This is notable when comparing patients who underwent surgery alone compared with historic RT approaches using preoperative chemoradiation (CRT).¹⁶ Therefore, minimizing CRT to only cases in which it is necessary is an important priority. We must also recognize that much of the morbidity arose from patients treated with historic RT techniques, and such toxicity may not be present with real-time adaptation using novel methods of image guidance. Yet, this remains to be proven. Figure 2 shows an example of an adaptive magnetic resonance (MR)–guided dose distribution that is adjusted daily for individual patient anatomy allowing for precise conformal and adapted RT dose distributions.^18,19 Such novel RT techniques must be tested prospectively in patients with rectal cancer as they offer significant potential advantages over historic approaches.

FIG 2.

Real-time MR-guided RT image using a 1.5-T MR linear accelerator. Axial image showing adaptive MR-guided boost from a prospective trial. Conformality of both the high- and intermediate-dose volumes can be seen. RT, radiation; MR, magnetic resonance.

SUMMARY OF THE RELEVANT LITERATURE

Rectal cancer has been the subject of multiple randomized trials and decades of research.²⁰ These data have resulted in a considerable prospective literature on the basis of which we can formulate decisions related to optimal patient management. It also creates a high benchmark standard from which expected outcomes are measured. Such a history is critical to consider, as we contemplate evolving areas of research that pertain to the challenges of curing rectal cancer patients. Two important areas that centrally relate to the CCTG C0.28 trial published in the Journal of Clinical Oncology²¹ include methods of OP with less extensive surgery and OP without surgery.

OP WITH LESS EXTENSIVE SURGERY

The concept of less extensive surgery consisting of LE after neoadjuvant therapy is decades old. To date, there have been a total of over 20 studies, mostly single-arm, which have examined this question.²² What remains as the most important end points in such trials is a robust focus on disease-free survival (DFS), rates of salvage TME, and functional QoL. Pure downstaging is perhaps less relevant than other end points specifically when compared with long-term oncologic control and QoL. The concept of administering some form of neoadjuvant therapy to shrink a tumor followed by a less extensive surgical resection is appealing, but needs to be approached with caution as oncologic outcomes could be compromised.

There have recently been several trials of neoadjuvant therapy followed by LE. A few principles should be considered when contemplating the prior and current relevant literature. First, accurate initial staging before administration of any local therapy becomes critical. Second is consideration of the number of patients included in the study and the adequacy of follow-up. Proper follow-up is essential to ensure the absence of local or regional recurrence events. Finally, a focus on QoL must be central to the examination of the findings.

The American College of Surgeons Oncology Group (ACOSOG) Z6041trial was a single-arm, open-label, phase II study focused on OP for clinical T2N0 distal rectal cancer. This trial used neoadjuvant concurrent CRT, giving standard doses (50.4-54 Gy at 1.8-2.0 Gy per fraction) with concurrent chemotherapy. Patients were clinically T2N0, staged by endorectal ultrasound or MRI using an endorectal coil. DFS was the primary end point, and evidence of local recurrence, distant metastasis, or death from any cause within 3 years counted as events. Impressively, the 3-year DFS in the intention-to-treat group was 88.2% and in the per-protocol group was 86.9%. These data indicate that a very high proportion of patients in this study could avoid transabdominal rectal resection without disease recurrence.²³ These promising results also were complimented by a high rate of functional preservation and QoL, with a total of 87% of patients completing both the Fecal Incontinence Severity Index (FISI) and the Functional Assessment of Cancer Therapy—Colorectal (FACT-C) questionnaires 1 year after surgery with no substantial deterioration in the overall FISI score.

A second important study recently published is the CRT therapy for rectal cancer in the distal rectum followed by organ-sparing transanal endoscopic microsurgery (CARTS) trial.²⁴ The results of this trial were less impressive than those of the ACOSOG Z6041 trial. Patients were included with T1-3N0 disease, which represented an important difference between the ACOSOG trial, which only included T1-T2N0. Neoadjuvant CRT was given to a total dose of 50 Gy over 25 fractions in the CARTS trial, and patients underwent TME 8-10 weeks after CRT. This was a multicenter trial performed at 12 hospitals. In total, two grade 5 toxicity events occurred during CRT in this trial. The presence of grade 5 toxicity events is concerning. Moreover, grade 3 toxicity events were seemingly higher than other rectal RT series.²³ In addition, the OP rates were lower in the CARTS trial in comparison with the ACOSOG Z6041 trial; however, such cross-trial comparisons are difficult, leading to important differences in the inclusion criteria.

There have been multiple other small and single-arm studies examining similar questions of neoadjuvant chemotherapy and RT before LE. A recently published systematic review demonstrated that of nine studies included, there is a pooled 3-year DFS rate of 92.8% and an impressive five-year DFS rate of 91.3%.²⁵ Although single-arm trials can provide helpful outcomes data, they are limited with regard to the ability to accurately compare across cohorts. Random assignment presents a critical opportunity to more accurately evaluate both oncologic outcomes and QoL between two interventions.

The GRECCAR2 trial was a prospective, randomized, open-label, multicenter, phase III trial at 15 tertiary centers in France.²⁶ This trial randomly assigned patients after chemoradiotherapy to either LE or TME. Grade 3 toxic effects were present in 15 (10%) of 145 patients. At 2 years, in the modified ITT population, an event from the composite primary outcome occurred in 41 (56%) of 73 patients in the LE group and 33 (48%) of 69 in the TME group (P = .43). In the modified ITT analysis, there was no difference between the groups in all components of the composite outcome, and superiority was not shown for LE over TME. Critical lessons were learned from the GRECCAR2 trial, specifically that composite end points may limit the ability to draw specific conclusions about QoL or oncologic end points. Moreover, patients have important preference, with approximately 15 percent of patients assigned to TME who refused radical surgery. It also confirmed that LE is oncologically safe after neoadjuvant therapy.

With this background, we approach the results of the recently published phase II NEO trial Canadian Cancer Trials Group (CCTG CO.28) published in this issue. In this trial, the historic use of neoadjuvant CRT, followed by LE, is replaced with 3 months of neoadjuvant FOLFOX (folinic acid/leucovorin-fluorouracil-oxaliplatin)/CAPOX (capecitabine-oxaliplatin), excluding RT entirely. This resulted in 79% of patients pursuing an organ-sparing approach, at the relatively early median follow-up point of 15.4 months. Similar to the ACOSOG Z6041 trial, patients had minimal QoL and functional score declines after chemotherapy and LE. Notably, rates of grade 3 and 4 toxicity were not insignificant on the CCTG CO.28 trial, with 36% of patients experiencing a grade 3 toxicity. Of the 58 patients on the trial, 56 underwent transanal excision surgery after chemotherapy with no unexpected toxicities and 12 (21%) underwent subsequent TME or CRT. Importantly, seven of the patients who underwent a TME had a pathologic complete response (pCR, ypT0N0). This result highlights our current inability to determine a near-complete response from a complete response using modern clinical examination tools including digital examination, endoscopy, and MRI. It also highlights the fact that a true nonoperative approach after chemotherapy without use of LE was not an option in this trial. Exploratory biomarker analysis by using next-generation sequencing revealed no association between RAS mutation status and tumor downstaging with chemotherapy. The authors rightly conclude that the findings from their study warrant further investigation. They also acknowledge multiple important limitations including the small size of the study and concentration in high-volume centers. Both aspects of the trial contribute to the necessity of further confirmatory studies before such an approach is offered outside of a trial.

OP WITH NONOPERATIVE MANAGEMENT

In the modern era, foregoing operative management of rectal adenocarcinoma through the use of RT and chemotherapy has been an emerging concept for nearly two decades.^27-33 Despite increasing patient and clinician interest, large-scale prospective, multi-institutional trials studying the safety and efficacy of integrating a nonoperative management (NOM) approach were lacking until recently.³⁴ OP with the use of chemotherapy and RT for stage II/III rectal cancer is now accepted by several international consensus guidelines. Recently, it has been included as an option in the national comprehensive cancer center (NCCN) guidelines.¹ Specifically, the idea of observing patients who achieve a complete response to neoadjuvant therapy is appealing on many levels as patients can avoid or defer life-altering surgery. Consequently, risks associated with surgical intervention, such as potential mortality, anastomotic dehiscence, wound complications, bleeding, or infection, and the chance for a temporary or permanent stoma are reduced or eliminated. In patients who require lifelong colostomies after an APR, such an operation can be anatomically disfiguring and upsetting. Avoiding these issues is of course appealing to patients. However, although NOM may sound appealing, it is not without the distinct possibility of long-term toxicity and functional detriment. Specifically, toxicity is still present with the use of chemotherapy and RT in rectal adenocarcinoma. However, the long-term toxicity after chemotherapy and RT is different than the functional detriment seen after surgery. Studying bowel function in this patient cohort through robust comparative, randomized, studies is important, but remains challenging, given the inherent limitations of the current tools used to measure patient-reported outcomes. Prospective studies need to better integrate and study QoL measures after an organ-sparing approach, compared with LE, or TME. Such research represents a critical and unmet clinical need. Recently, consensus end points to optimally evaluate outcomes from different approaches have been identified.³⁵ Phase II and (if feasible) subsequent phase III randomized trials examining distinct approaches to accomplish OP are needed.

There have been numerous retrospective series and multi-institutional cohorts that have examined the outcomes associated with NOM of rectal adenocarcinoma.^27-33,36 Several series have published outcomes associated with OP. Yet, there remains only a few relatively small prospective trials addressing this topic. More importantly, there are a variety of methods by which OP can be accomplished. Several series from Europe have accomplished impressive rates of OP using high-dose focused CRT alone, either with brachytherapy or image-guided intensity-modulated RT.^37,38 Interestingly, these series did not use systemic chemotherapy. Such approaches warrant potential comparison with longer courses of systemic chemotherapy. For patients with cT1-T3b rectal tumors with preserved mismatch repair, there are indeed multiple existing options including (1) neoadjuvant CRT followed by LE, (2) dose-escalated CRT (3) TNT including systemic chemotherapy and CRT followed by observation, and, now with the CCTG CO.28 trial, (4) neoadjuvant chemotherapy (without RT) followed by LE.³⁷ Limitations of the current prospective data are the absence of random assignment, and little (if any) focus on functional QoL differences across the current available organ-sparing strategies. Never before has there been a greater need for appropriately powered, randomized studies that will correctly examine the potential differences in patient-reported QoL outcomes. Recently published prospective OP trials are summarized in Table 1.

TABLE 1.

Prospective, Nonoperative Management Studies in Rectal Cancer

SUGGESTED APPROACHES TO MANAGEMENT

Changes to rectal cancer management must take place cautiously, especially as data are rapidly evolving. If centers are embarking on NOM, using either CRT or LE after neoadjuvant therapy, it should be acknowledged that nearly all prospective trials examining the safety and efficacy of this technique have taken place in high-volume centers.^34,37,39 Experience in staging, optimal chemotherapy dosing, conformal RT, and precise evaluation of response or candidacy for LE require significant multidisciplinary experience. Patient preference and shared decision making must also be accounted for while keeping the patient at the center of the discussion. It is clear that some patients have strong desires for OP, while others have no strong preferences and look to their oncologists to help with such decision making.⁴⁰ Currently, progression-free survival and overall survival outcomes are excellent for patients with rectal cancer, and prospective data reflecting the ability to achieve high rates of TME-free survival with TNT are emerging. Small studies, with limited follow-up, such as that presented by Kennecke et al in the CCTG CO.28 trial, although important for progress in the field, should not be considered as a standard of care at this time. Instead, these studies represent important data to guide future studies and to better evaluate optimal NOM applications in rectal cancer. Future trials could also include comparisons with other strategies including RT such as that used in ACOSOG Z6041.

The future of rectal cancer management will depend on thoughtful, patient-centered, and creative clinical trial design. Focus should start to shift from progression-free survival and overall survival, to better integration of clinical complete response (cCR), QoL, and patient preference-related outcomes. It is important to consider that improvements in RT techniques (such as adaptive MR-guided boosting) represent very exciting areas with minimal data having examined safety and efficacy, although small prospective trials are ongoing (ClinicalTrials.gov identifier: NCT04808323).¹⁸ Despite significant capabilities with modern technology to modulate RT dose, the majority of published or ongoing studies have used historic (even antiquated) RT doses and techniques of delivery that have remained unchanged for decades. Moreover, new subtypes of rectal cancer will emerge and demonstrate exquisite responses to targeted, optimal neoadjuvant treatment strategies allowing more patients the opportunity to be considered as candidates for OP. For example, in the 5%-8% of patients with mismatch repair deficient rectal cancer, use of PD-1 inhibition has led to a remarkable clinical complete response rate to date and opened the door to long-term OP, yet this dramatic clinical response after PD-1 blockade will require expansion of the patient cohort and longer surveillance to test durability.⁴¹

The standard of cancer for rectal cancer management is changing rapidly. Today, patients desiring OP are best served by enrollment in a clinical trial. Those with stage II/III pMMR tumors ineligible for a trial, or who decline clinical trial enrollment, should be treated with TNT including CRT first and systemic chemotherapy.³⁴ This standard is largely on the basis of the existing data relative to the outcomes desired, the techniques used, and consensus guidelines.¹ Patients should be managed in high-volume and experienced centers, and also educated that OP is a new standard for which long-term safety and efficacy data are emerging.

PATIENT MANAGEMENT

Tumor board consensus was for surgical resection. An LAR with a colonic J-pouch and diverting loop ileostomy without complications was performed. Final pathology showed invasive adenocarcinoma arising in a tubulovillous adenoma. The resection margins were negative and 23 lymph nodes were negative for carcinoma (0/23). No adjuvant therapy was recommended. The patient recovered from surgery and has excellent bowel function.

AUTHOR CONTRIBUTIONS

Conception and design: All authors

Collection and assembly of data: All authors

Data analysis and interpretation: All authors

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Achieving a Cure Without Total Mesorectal Excision in Rectal Adenocarcinoma

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).