Treatment of locally recurrent rectal adenocarcinoma

INTRODUCTION — Surgical resection is the cornerstone of curative therapy for patients with potentially resectable rectal cancer. Radiation therapy with concurrent fluoropyrimidine chemotherapy has emerged as an important component of curative therapy for transmural or node-positive rectal cancer because in contrast to colon cancer, in which the failure pattern is predominantly distant, the site of first failure can be local as well as distant.

Chemoradiotherapy is often administered preoperatively for clinically staged T3 or T4 (table 1), or node-positive tumors; for distal tumors, in which tumor regression may allow successful conversion of a planned abdominoperineal resection to a sphincter-sparing surgical procedure; or if the preoperative staging evaluation suggests mesorectal fascia invasion. Preoperative, as compared with postoperative, chemoradiotherapy results in a superior sphincter preservation rate, a lower rate of anastomotic stenosis, and better local control, while providing similar long-term survival. Good-quality surgery (total mesorectal excision [TME]) is also associated with lower local failure rates. (See "Neoadjuvant therapy for rectal adenocarcinoma", section on 'Indications for neoadjuvant treatment' and "Radical resection of rectal cancer", section on 'Total mesorectal excision'.)

Despite refinements in surgical techniques and optimal use of neoadjuvant and adjuvant therapies, the incidence of locoregional relapse after treatment of locally advanced rectal cancer is still 4 to 8 percent [1-3]; nevertheless, distant recurrences still predominate [3-6].

Treatment of locally recurrent rectal cancer will be discussed here. Neoadjuvant chemoradiotherapy for potentially resectable primary adenocarcinomas, adjuvant therapy after resection of a primary rectal adenocarcinoma, pretreatment local staging evaluation, surgical principles, and recommendations for posttreatment surveillance are discussed separately, as is the management of rectal squamous cell cancers. (See "Neoadjuvant therapy for rectal adenocarcinoma" and "Pretreatment local staging evaluation for rectal cancer" and "Radical resection of rectal cancer" and "Adjuvant therapy for resected rectal adenocarcinoma in patients not receiving neoadjuvant therapy" and "Post-treatment surveillance after colorectal cancer treatment" and "Treatment of anal cancer", section on 'Rectal squamous cell cancers'.)

THE CHALLENGE OF LOCALLY RECURRENT RECTAL CANCER — Management of locally recurrent rectal adenocarcinoma is a significant challenge for a number of reasons:

●Patients with locally recurrent rectal cancer form a very heterogeneous group, ranging from focused anastomotic recurrence to the involvement of adjacent soft tissue organs and the bony structures of the pelvis. Without treatment, these patients have a short life expectancy that is often complicated by relentless pelvic pain, malodorous discharge, and uncontrollable tenesmus, which significantly impairs quality of life [7].

●The clinical nature and prognosis of locally recurrent rectal cancer have changed since the introduction of preoperative radiation therapy (RT):

RT reduces local recurrence at all subsites but is especially effective at preventing anastomotic recurrences [8]. As a result, most recurrences after pelvic RT are not at the anastomotic site, and surgical treatment of nonanastomotic recurrences is more difficult.

Many patients who present with a local recurrence after prior RT have simultaneous distant metastases [9-12]. As an example, in the Dutch total mesorectal excision (TME) trial, 83 of the 129 patients with a local recurrence also had distant metastases (63 percent) [9]. One systematic review concluded that only approximately 40 percent of patients with locally recurrent rectal cancer are candidates for potentially curative treatment [13].

In part related to these issues, long-term outcomes seem to be worse after local recurrence in previously irradiated patients as compared with in patients treated initially with surgery alone [8,12,14].

●The choice of therapy must take into consideration prior therapy, the local extent of the recurrence, and whether or not distant disease is present. For highly selected patients, surgery alone may be associated with long-term survival.

However, for most patients with an isolated local recurrence, combined modality therapy may provide better outcomes, particularly if patients have not been previously irradiated. Patients with distant metastases are usually approached initially with systemic chemotherapy to allow the natural history of the recurrent disease to declare itself. Some of these patients may eventually be determined to have potentially resectable distant (eg, liver, lung) as well as locally recurrent disease. For others who have unresectable distant metastases, palliation of pain, obstruction, or bleeding caused by a locally recurrent primary tumor may usually be achieved through a diverting colostomy or by nonsurgical means, such as with an endoscopically placed, intraluminal, self-expanding metal stent (for tumors not in the mid and distal rectum). (See "Systemic therapy for nonoperable metastatic colorectal cancer: Selecting the initial therapeutic approach" and "Locoregional methods for management and palliation in patients who present with stage IV colorectal cancer", section on 'Management of the primary cancer' and "Enteral stents for the management of malignant colorectal obstruction".)

MODE OF PRESENTATION — Approximately one-third of patients may be asymptomatic at the time of recurrence, with the recurrence discovered during a routine postoperative follow-up evaluation [15-18]. (See "Post-treatment surveillance after colorectal cancer treatment".)

Anastomotic recurrence can present with rectal bleeding or alteration of bowel habits. Pain is a more concerning symptom that may be associated with involvement or compression of organs, bone, or nerves.

Local recurrence after an abdominoperineal resection can present with a nonhealing perineal wound, a perineal mass, pelvic pain, or bowel obstruction that results from involvement of small bowel in the pelvic mass.

Up to 74 percent of patients with locally recurrent rectal cancer after combined modality therapy will present with synchronous distant metastatic disease [9-13,19,20].

PRETREATMENT EVALUATION — The pretreatment evaluation should focus on assessment of fitness for major surgical intervention, and staging of the tumor to ascertain the anatomy and the extent of local and distant disease. Locally recurrent rectal cancer is challenging to manage, particularly in patients who have received prior radiation therapy, and referral to a multidisciplinary team of specialists at a center with expertise in managing these complex cases is preferred. Frequently, input from experts in colorectal surgery, urology, plastic surgery, orthopedic oncology, neurosurgery, medical oncology, and radiation oncology is needed to develop an optimal approach.

All patients with suspected locally recurrent rectal cancer should undergo a full clinical staging evaluation to exclude the presence of distant metastatic disease, including a computed tomography (CT) of the chest, abdomen, and pelvis and an integrated positron emission tomography (PET)/CT scan. Locally recurrent or metastatic disease should be confirmed by biopsy.

In addition to a digital clinical examination (which may require examination under anesthesia) and full colonoscopy to evaluate the anastomotic site and assess the remainder of the large bowel for synchronous lesions, the preoperative evaluation of any patient with a locally recurrent rectal cancer should also include pelvic magnetic resonance imaging (MRI) [21-23]. In general, endorectal ultrasound is a less helpful modality than MRI for staging a local recurrence after prior surgical resection.

For women with a suspected anterior recurrence, a vaginal examination must be performed; rarely, cystoscopy is needed in the presence of hematuria or if the scans raise the possibility of bladder invasion. In addition, serum levels of the tumor marker carcinoembryonic antigen (CEA) should be obtained. (See "Clinical presentation, diagnosis, and staging of colorectal cancer", section on 'Clinical staging evaluation' and "Pretreatment local staging evaluation for rectal cancer", section on 'Imaging evaluation'.)

Despite technical advances, preoperative imaging, particularly CT scan, tends to underestimate the extent of local infiltration and adjacent organ involvement by the tumor. The final decision for resectability of a local recurrence is often made intraoperatively, with consideration of the location of the recurrence and the presence of osseous, adjacent organ, vascular, lymphatic, and/or nerve root involvement.

Another important point is that on pretreatment imaging, it can be difficult to distinguish between soft tissue from ongoing inflammation that is mildly fludeoxyglucose (FDG) avid and actual tumor recurrence. Biopsies of this area can be difficult, and a negative result may not be reliable. Sometimes, serial imaging is needed to distinguish between fibrosis and recurrent cancer.

Biopsy — Histologic confirmation of the recurrence is an important step before proceeding with extensive radical surgery, especially if induction chemoradiotherapy is planned. This may be accomplished by colonoscopy for an intraluminal recurrence. When the tumor is not accessible by endoscopy, CT-guided biopsy can often be performed.

When a tissue diagnosis cannot be obtained (eg, as a result of the location of the recurrence or if there is a significant degree of surrounding fibrosis) and there is convincing radiologic evidence of recurrence (eg, highly suggestive PET scan or serial radiologic examinations that demonstrate progressive enlargement of a suspected recurrent mass), it is reasonable to pursue multimodality treatment [24].

Classification of recurrence pattern — Preoperative imaging studies and operative findings are used to categorize the recurrence. There is no universal consensus on the best way to classify locally recurrent rectal cancer for the purpose of prognostication and treatment selection. Several have been proposed:

●A system proposed by the Mayo Clinic classifies patients into groups based on the presence of pain (S0 = asymptomatic; S1 = symptomatic without pain; S2 = symptomatic with pain) and the degree of fixation to anatomical structures located at the anterior, sacral, left, and right sides of the pelvis (F0 = not fixed to any site; F1 = fixed to one site; F2 = fixed at two sites; F3 = fixed at three sites) [25]. While the degree of fixation and pain did not appear to adversely affect survival in their initial report, a subsequent series with more patients indicated significantly reduced survival with symptomatic pain and more than one point of fixation [16].

●A different classification system proposed by the Memorial Sloan Kettering Cancer Center (MSKCC) was based upon the anatomical location of the recurrence within the pelvis and the structures that were involved [26]. Recurrences were defined as one of the following:

•Axial – Anastomotic recurrence after low anterior resection, perineal recurrence after abdominoperineal resection, or local recurrence after transanal excision

•Anterior – Involving urologic or gynecologic organs

•Posterior – Involving the sacrum or coccyx

•Lateral – Involving the pelvic sidewall structures

Using this system, tumors with a lateral component had a significantly reduced likelihood of a microscopically complete (R0) resection.

●Another classification of patterns of invasion was proposed by the Leeds group in 2005 [27]:

•Central – Tumor confined to pelvic organs or connective tissue without contact or invasion into bone

•Sacral – Tumor present in the presacral space and abuts into or invades into the sacrum

•Sidewall – Tumor involving the structures on the lateral pelvic sidewall, including the greater sciatic foramen, and the sciatic nerve through to the piriformis and the gluteal region

•Composite – Sacral and sidewall recurrence combined

This group did not address the prognostic or therapeutic implications of the classification system; however, they did provide an algorithmic approach to surgical management based upon the location of the recurrence.

MANAGEMENT

Surgical resection — Complete radical resection is a prerequisite for cure; outcomes are superior if microscopically complete (R0) resection can be achieved, even if this requires an extensive operative procedure (eg, pelvic exenteration) [28-32]. In these settings, the benefits of surgery (potential for cure, improved symptoms, quality of life) must be weighed against the postoperative morbidity, possibility of multiple permanent stomas, chronic pain, and even the limited survival that result from aggressive surgery.

Complete surgical removal offers the only hope for long-term survival in locally recurrent rectal cancer. External beam radiation therapy (RT) and/or chemotherapy alone may successfully palliate symptoms and prolong median survival to 12 to 15 months (compared with a median survival of three to eight months for untreated local recurrence), but five-year survival rates are <5 percent [17,33-36].

At some institutions, surgical resection alone is pursued for a small subset of patients with an isolated, small, central recurrence at the anastomotic site given some reports of long-term survival if the recurrence can be completely resected with negative margins [15,37-44]. However, others note that oncologic outcomes are poor, even in highly selected patients who undergo surgery for an anastomotic recurrence after transanal excision of an early stage rectal cancer [45,46]. In our view, surgery alone should not be utilized in the management of a local recurrence, particularly if RT has not been previously utilized. (See 'Patients with no prior radiation therapy' below.)

Even if it has been given within the past year, intraoperative radiation therapy (IORT) may still be an option [47]. (See 'Role of intraoperative radiation therapy' below.)

Aggressive attempts to obtain microscopically negative margins are warranted because a negative-margin resection is the single most important predictor of cancer-specific survival in patients undergoing surgery for recurrent rectal cancer [27,48-52].

For more extensive recurrences, a more extensive surgical procedure (eg, posterior or total pelvic exenteration for an anteriorly located recurrence, abdominosacral resection or partial sacrectomy for a posteriorly located recurrence) may be required. In these settings, the benefits of surgery (potential for cure, improved symptoms, quality of life) must be weighed against the surgical morbidity, possibility of multiple permanent stomas, chronic pain, and even limited survival that result from aggressive surgery. (See "Surgical treatment of rectal cancer", section on 'Multivisceral resection'.)

Resectability in recurrent rectal cancer is defined as the ability to complete a surgical resection with an R0 margin and acceptable postoperative morbidity and mortality. Tumor location and degree of local invasion are the most important factors in determining resectability. Contraindications to radical surgery include [24,26]:

●Nerve root involvement at or above the level of S1-2

●Proximal (S1, S2) sacral invasion extending to the sacral promontory (relative contraindication)

●Involvement of the paraaortic lymph nodes

●Tumor encasement of the external iliac vessels

●Extension of tumor through the greater sciatic notch

●Bilateral ureteral obstruction (relative contraindication)

●Unresectable extrapelvic disease

●Circumferential involvement of the pelvic wall

In general, these factors preclude a potentially curative surgical procedure. However, there are reports in which extensive sacropelvic resections as well as hemipelvectomies have been performed successfully [37,53-56]. In general, these are very highly selected patients at institutions experienced in these types of procedures, and perioperative morbidity rates are high.

●At the University of Sydney, 49 patients underwent extended resections (defined as higher than the S3 spinal level) for locally recurrent disease. These procedures included 37 total pelvic exenterations [54]. Sixteen patients also had other bony structures resected aside from the sacrum. There were 21 high sacral resections, defined as resection of the sacrum proximal to S3. R0 resection margins were achieved in 74 percent of cases. There were no postoperative mortalities. However, there were 39 percent major morbidity and 78 percent minor morbidity in 40 of the 49 patients (82 percent overall morbidity). Eight of 10 patients resected at the S1 spinal level experienced major morbidity. The estimated five-year disease-free survival was 43 percent in those resected with R0 margins. There was a 45 percent relapse rate. The median overall survival for a patient undergoing R0 resection was 59 months, while it was 28 months for those undergoing R1 (microscopically positive) or R2 (gross residual disease) resection. In this series, six patients developed sensory loss, and seven patients experienced motor deficit postoperatively.

Distant metastases (typically to the liver or lung) may or may not be a relative contraindication to radical surgery depending on the location and potential for curative resection. (See "Potentially resectable colorectal cancer liver metastases: Integration of surgery and chemotherapy" and "Surgical resection of pulmonary metastases: Outcomes by histology".)

Preoperative imaging is one of the most important tools the surgeon has to select patients for these procedures. However, despite technical advances, preoperative imaging, particularly computed tomography (CT) scan, tends to underestimate the extent of local infiltration and adjacent organ involvement by the tumor. The final decision for resectability is often made intraoperatively, with consideration of the location of the recurrence and the presence of osseous, adjacent organ, vascular, lymphatic, and/or nerve root involvement.

Although radical pelvic resection has traditionally been associated with the need for permanent fecal and urinary diversion, continence preservation may be feasible if an oncologically sound resection of the recurrent tumor can be carried out without impairing the integrity of the pelvic floor muscles [57,58]. Ideally, the management of these patients should be by an experienced multidisciplinary surgical team, which may include colorectal surgeons (or surgical oncologists), plastic and reconstructive surgeons, urologists, and neurosurgeons, among others. (See "Surgical treatment of rectal cancer", section on 'Multivisceral resection'.)

Combined modality therapy — For most cases of locally recurrent rectal cancer, we suggest combined modality therapy rather than surgery alone. The specific approach depends on whether the patient received prior RT or not.

There are no randomized trials comparing surgery alone with combined modality therapy, either in patients who have or have not received prior RT. Most retrospective series of patients treated for locally recurrent rectal cancer include a mix of patients with and without prior RT. In general, most reports show better outcomes, including higher resectability rates, in patients treated with a combined modality approach as compared with surgery alone, regardless of the use of prior RT [15,31,47,49,52,59-62].

The largest series was a pooled analysis of 565 patients treated for locally recurrent rectal cancer at one of two major treatment centers (one Dutch, one American) [31]. Overall, 251 (44 percent) had a radical (R0) resection, and complete resection rates were higher among those who received preoperative treatment (43 and 50 percent for chemoradiotherapy and full-course RT, respectively, compared with 26 percent for those without any preoperative treatment). Patients who were reirradiated had a similar outcome compared with patients who were RT-naïve and could undergo full-course RT, except that they had a higher local recurrence rate. In multivariate analysis, R0 resection and preoperative therapy had an impact on both local recurrence and overall survival.

Patients with no prior radiation therapy — For previously unirradiated patients with an isolated local recurrence, we follow the same principles as guide preoperative therapy of primary locally advanced rectal adenocarcinoma. For most patients, we recommend preoperative therapy with fluoropyrimidine-based, concurrent, long-course external beam RT, followed by surgery 8 to 10 weeks later, and then four to six months of adjuvant chemotherapy. (See "Neoadjuvant therapy for rectal adenocarcinoma", section on 'Long-course chemoradiation'.)

Another option for patients with bulky recurrent tumors or extensive nodal disease is initial chemotherapy with an oxaliplatin-based chemotherapy regimen (eg, oxaliplatin plus leucovorin and short-term infusional fluorouracil [FOLFOX], or oxaliplatin plus capecitabine [CAPOX], unless there is a contraindication to the use of oxaliplatin) for up to four months in responding patients [63,64]. Given the importance of R0 resection, we often consider more aggressive chemotherapy (eg, oxaliplatin plus irinotecan, leucovorin, and short-term infusional fluorouracil [FOLFOXIRI] (table 2)) if we are trying to maximize the objective response rate to chemotherapy and the patient can tolerate it. Chemotherapy is then followed by long-course chemoradiotherapy and then resection. (See "Neoadjuvant therapy for rectal adenocarcinoma", section on 'Total neoadjuvant therapy for locally advanced tumors' and "Treatment protocols for small and large bowel cancer".)

Another option is short-course RT followed by fluoropyrimidine and oxaliplatin-based chemotherapy [65]. (See "Neoadjuvant therapy for rectal adenocarcinoma", section on 'Short-course radiotherapy'.)

Outcomes — The efficacy of concurrent preoperative chemoradiotherapy for locally recurrent rectal cancer in previously unirradiated patients is supported by data from retrospective series [16,18,27,52,66-70] and a single randomized trial of preoperative chemoradiotherapy versus RT alone that included patients with locally recurrent disease [71]. There are no randomized trials of resection alone versus combined modality therapy.

In several retrospective reports, rates of R0 resection in patients undergoing multimodality treatment range from 32 to 82 percent, five-year overall survival rates are 20 to 40 percent (37 to 60 percent among those undergoing R0 resection), perioperative morbidity and mortality rates are 24 to 44 and 0.6 to 9 percent, respectively, and rates of distant recurrence or a second local recurrence are 50 and 38 to 50 percent, respectively [16-18,21,27,66,70].

In the randomized trial, 207 patients with unresectable primary (n = 182) or locally recurrent (n = 25) rectal cancer were randomly assigned to RT (50 Gy) with or without concomitant chemotherapy with bolus fluorouracil [71]. Additional chemotherapy (leucovorin-modulated fluorouracil) was given for 16 weeks after surgery as adjuvant chemotherapy. None of the patients with locally recurrent tumors had been previously irradiated.

In the entire population, compared with preoperative RT alone, preoperative chemoradiotherapy was associated with a significantly higher R0 resection rate (84 versus 68 percent) and pathologic complete response rate (16 versus 7 percent), as well as better five-year rates of local control (82 versus 67 percent), failure-free survival (63 versus 44 percent), and cancer-specific survival (72 versus 55 percent). Results were not stratified according to primary versus recurrent cancer. However, as a group, patients treated for locally recurrent cancer (n = 25) had a worse outcome than did those with primary disease (five-year failure-free and overall survival 30 and 37 percent, respectively). Furthermore, these patients were more likely to be categorically unresectable following preoperative therapy.

Treatment intensification increases toxicity and may not improve outcomes. This was shown in a retrospective report of 102 patients with locally recurrent rectal cancer after surgery alone who were treated with preoperative long-course RT (50.4 Gy) concurrent with FOLFOX given every two weeks for 12 weeks [32]. Ninety-six completed chemoradiotherapy, of whom 75 went on to surgery (low anterior resection in 33, abdominal perineal resection in 23, Hartmann procedure in six, and pelvic exenteration in 13). There were 52 R0 resections, and the remainder had microscopic residual disease (13) or gross disease left behind (10). There were two perioperative deaths (both from sepsis) and 20 surgical complications (20 percent), which included four anastomotic leakages, three pelvic abscesses, eight wound infections, three intestinal obstructions, and two voiding dysfunctions. At a median follow-up of 46 months, the five-year overall and disease-free survival rates were 45 and 14 percent, respectively.

Prognostic factors after combined modality therapy were addressed in a series of 94 patients who received RT for previously unirradiated locally recurrent rectal cancer; 63 also received chemotherapy, and 46 had resection [67]. At a median follow-up of 19 months, tumor recurred locally in 49 (52 percent) and with distant metastases in 57 (61 percent). For the entire cohort, survival rates at one, two, and three years were 74, 56, and 36 percent, respectively. In multivariate analysis, factors that were associated with a significantly greater likelihood of survival were better performance status, lower American Joint Committee on Cancer (AJCC) stage at recurrence, and higher hemoglobin levels during RT.

Role of intraoperative radiation therapy — At many centers in the United States, Europe, and Asia, IORT has been used in conjunction with preoperative irradiation (with and without fluorouracil) and surgical resection when there is strong suspicion for gross residual cancer, positive resection margins, or the presence of tumor adherence to adjacent structures (eg, the pelvic sidewall or sacrum) [31,47,72-78]. IORT is usually not given if at the time of surgery, there is no tumor adherence and adequate soft tissue radial margins are present (>1 cm).

Although there are no randomized trials in which high-risk patients were randomized to receive or not receive IORT, favorable results are reported in series that utilize combined modality therapy that includes IORT administered at the time of surgical resection for locally recurrent disease; in most cases, IORT was preceded by external beam RT, usually with a concurrent fluoropyrimidine [19,31,47,78-83]. The largest series was a pooled analysis of 565 patients treated for locally recurrent rectal cancer at one of two major treatment centers (one Dutch, one American); all but 12 received IORT [31]. Overall, 251 (44 percent) had a radical (R0) resection, and the main factors related to the radicality of resection were preoperative treatment and a longer waiting time from the end of preoperative therapy to surgical resection with IORT. Radical resection was the most important factor influencing all oncologic outcomes. Overall survival rates at three and five years by extent of resection were R0 66 and 48 percent, R1 47 and 25 percent, and R2 37 and 17 percent, respectively.

An important point is that the use of IORT cannot compensate for incomplete resection in such patients [52,81,84]. This was illustrated in a series of 41 patients with locally recurrent rectal or rectosigmoid cancer who were treated with preoperative external beam RT, resection, and IORT [84]. At five years, the local control and survival rates for patients undergoing complete surgical resection were 47 and 21 percent, respectively, while the corresponding values for those undergoing only subtotal resection were 21 and 7 percent, despite the use of IORT in all patients.

Typical IORT doses are in the range of 10 to 20 Gy. The IORT dose is usually 10 to 12.5 Gy in patients undergoing complete resection with negative margins, 12.5 to 15 Gy in patients undergoing subtotal resection with microscopic residual, and 17.5 to 20 Gy in patients with macroscopic tumor after resection.

When IORT is used, it is critical for the surgeon and radiation oncologist to define all high-risk areas to determine the optimal position for the IORT field. To direct the IORT, cones are used with internal diameters ranging from 4 to 8 cm; cone size is selected to fully cover the high-risk area, generally on the sacrum or pelvic sidewall. The cone must abut the site being treated, which can be difficult if the high-risk area is located in an anatomically confined region such as the pelvis. Most IORT treatments in rectal cancer are given through the abdomen, but a perineal port is occasionally used to treat very low-lying tumor involving the coccyx, distal pelvic side wall, or portions of the prostate and bladder when an exenteration is not performed. IORT can also be delivered using high-dose-rate brachytherapy using a Harrison-Anderson-Mick (HAM) applicator.

Previously irradiated patients — For previously irradiated patients, pelvic reirradiation is feasible in selected patients and may permit surgical salvage and long-term survival. Patients being considered for reirradiation need to be carefully selected, and careful attention to dose and technique is required to minimize the risk of serious late effects. Sophisticated RT techniques, such as intensity-modulated RT or proton beam irradiation, are required to stay within normal tissue tolerance constraints. The dose of RT for reirradiation in this setting should be limited to 30 to 39 Gy.

Surgery alone may result in long-term disease control in some patients who recur locally after prior combined modality therapy; however, outcomes are generally poor, and long-term cures are uncommon [3,14,48]. As an example, in one report of 50 patients who underwent surgical exploration for locally recurrent rectal cancer after primary combined modality therapy, only 24 could undergo an R0 resection (48 percent) [48]. Despite adjuvant chemotherapy in all patients, only eight (16 percent of the total) remained disease free for periods ranging from 21 to 69 months.

Pelvic reirradiation has generally not been undertaken for fear of prohibitive normal tissue complications. However, it is feasible in selected patients and may permit successful surgical salvage and long-term survival, albeit with higher complication rates than with reirradiation alone [52,85-90]. The following studies represent the range of findings:

●The benefits of reirradiation with or without surgery were assessed in a systematic review and meta-analysis of 17 studies (three prospective uncontrolled trials, the remainder retrospective reports) involving 744 patients [90]. For patients who underwent reirradiation plus surgery, pooled one-, two-, and three-year survival rates were 86, 72, and 52 percent, respectively, while for those undergoing reirradiation alone, the corresponding values were 64, 34, and 24 percent, respectively. The benefits of combined modality treatment came at the cost of more grade 3 or worse acute and late complications (12 versus 26 percent in the nonsurgical and surgically treated patients, respectively). In four studies that compared surgically and nonsurgically treated patients, those who underwent surgery had a sixfold higher risk of late complications (gastrointestinal, genitourinary, skin, and soft tissue).

●The importance of R0 resection was emphasized in a retrospective study from the Netherlands of 147 patients with locally recurrent rectal cancer, most of whom (n = 124) underwent preoperative combined modality treatment followed by extensive surgery and IORT; the remainder had surgery with or without postoperative adjuvant treatment [52]. Fifty-seven (39 percent) had undergone prior pelvic irradiation, and their course of RT was limited to 30.6 Gy; other patients received 50.4 Gy.

Patients who received either full-course irradiation or reirradiation as a component of treatment survived significantly longer and lasted longer without local recurrence or metastases than did those who were not irradiated. However, the single most important factor in predicting survival after radical surgery plus IORT for locally recurrent rectal cancer was a negative margin of resection. In the entire cohort, the actuarial five-year overall and cancer-specific survival rates were 32 and 39 percent, respectively, while among those with a R0 resection, the actuarial five-year overall and cancer-specific survival rates were 48 and 58 percent, respectively.

●Long-term toxicity to normal tissues depends in part upon the size of each treatment fraction as well as the total radiation dose. Decreasing the size of each radiation fraction should permit higher total doses without increasing late morbidity. Hyperfractionation administers multiple daily treatments with smaller-than-conventional fraction sizes given over approximately the same treatment duration.

The benefits for hyperfractionated RT in the setting of potentially curative treatment for previously irradiated recurrent rectal cancer have been shown in multiple retrospective reports and uncontrolled studies [86,87,91-93]. In a representative phase II study of 59 previously irradiated patients with an isolated local recurrence of rectal cancer, treatment consisted of hyperfractionated RT administered as 1.2 Gy fractions twice daily to a total dose of 30 Gy; concomitant chemotherapy (infusional fluorouracil) was administered to all patients [92]. Treatment was well tolerated overall; 51 of the 59 patients completed chemoradiotherapy without treatment interruption, while six had a temporary interruption and only two discontinued treatment permanently. Grade 3 or lower gastrointestinal toxicity developed during treatment in only 5 percent, and there was no grade 4 toxicity. Tumor resection was performed in 30 patients (51 percent) and was R0 in 21 (36 percent). Five-year actuarial survival was 39 percent, and it was 67 percent in R0-resected patients. Late toxicities were infrequent and generally mild, and included skin fibrosis, male impotence, urinary incontinence, small bowel obstruction, and dysuria in 2, 2, 1, 1, and 1 percent of treated patients, respectively.

On the other hand, the systematic review mentioned above did not report any significant differences in acute or late toxicity for patients who were reirradiated using hyperfractionation versus conventional fractionation RT [90].

Benefit of adjuvant/neoadjuvant chemotherapy — There are few data to guide the use of adjuvant chemotherapy after combined modality therapy of a locally recurrent rectal adenocarcinoma if it was not administered in the neoadjuvant (induction therapy) setting, and its role is not established. In the few retrospective reports that include the use of adjuvant chemotherapy as a component of combined modality therapy, none of the reports concludes that administration of adjuvant chemotherapy contributed to more favorable long-term outcomes [12,48,60,83,92]. We would generally only consider adjuvant chemotherapy in a patient who had not received treatment with an oxaliplatin-based chemotherapy regimen (eg, FOLFOX, CAPOX) after initial resection.

At many institutions, including that of the authors, patients with bulky recurrent tumors or extensive nodal disease are offered initial chemotherapy with an oxaliplatin-based chemotherapy regimen (eg, FOLFOX, CAPOX, or FOLFOXIRI, unless there is a contraindication to the use of oxaliplatin) for up to four months in responding patients [63,64]. We would not offer adjuvant chemotherapy to a patient who has received preoperative chemotherapy. (See "Adjuvant therapy for resected rectal adenocarcinoma in patients not receiving neoadjuvant therapy", section on 'Oxaliplatin versus fluoropyrimidines alone' and "Adjuvant therapy after neoadjuvant therapy for rectal cancer", section on 'Choice of postoperative regimen' and "Treatment protocols for small and large bowel cancer".)

PALLIATION OF OBSTRUCTIVE SYMPTOMS DUE TO LOCALLY ADVANCED DISEASE — Palliation should be considered in symptomatic patients who are not candidates for potentially curative multimodality therapy. Symptom relief can often be achieved through a diverting colostomy, endoscopic laser ablation, or stent placement (for tumors not in the mid or distal rectum); palliative tumor resection is uncommonly needed. This topic is discussed in detail separately. (See "Locoregional methods for management and palliation in patients who present with stage IV colorectal cancer" and "Enteral stents for the management of malignant colorectal obstruction".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Colorectal cancer".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Beyond the Basics topics (see "Patient education: Colon and rectal cancer (Beyond the Basics)" and "Patient education: Colorectal cancer treatment; metastatic cancer (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Management of locally recurrent rectal adenocarcinoma is a significant challenge. The choice of therapy depends upon prior therapy, the local extent of the recurrence, and whether or not distant disease is present. (See 'Introduction' above.)

●Approximately one-third of patients with a local recurrence are asymptomatic, and the recurrence is discovered during a routine postoperative follow-up evaluation. Many patients with locally recurrent rectal cancer after initial combined modality therapy have synchronous distant metastatic disease. (See 'Mode of presentation' above.)

●The pretreatment evaluation should focus on assessment of fitness for major surgical intervention, staging of the tumor to ascertain the anatomy and the extent of local and distant disease, and histologic confirmation of the recurrence. All patients with suspected locally recurrent rectal cancer should undergo a full clinical staging evaluation to exclude the presence of distant metastatic disease, including a computed tomography (CT) of the torso and an integrated positron emission tomography (PET)/CT scan. In addition to a digital clinical examination (which may require examination under anesthesia), a full colonoscopy, to evaluate the anastomotic site and assess the remainder of the large bowel for synchronous lesions, and rectal magnetic resonance imaging (MRI) are recommended. Gynecologic or cystoscopic evaluation may be needed for selected patients. In addition, serum levels of the tumor marker carcinoembryonic antigen (CEA) should be obtained. (See 'Pretreatment evaluation' above.)

●Complete radical resection is a prerequisite for cure. Aggressive surgical attempts to obtain microscopically negative margins are warranted due to the superior outcomes with microscopically complete (R0) resection. This may require an extensive operative procedure (eg, pelvic exenteration). (See 'Surgical resection' above.)

Contraindications for radical surgery include:

•Nerve root involvement above the level of S1-2

•Proximal (S1, S2) sacral invasion extending to the sacral promontory (relative contraindication)

•Involvement of the paraaortic lymph nodes

•Tumor encasement of the external iliac vessels

•Extension of tumor through the greater sciatic notch

•Bilateral ureteral obstruction (relative contraindication)

•Unresectable extrapelvic disease

•Circumferential involvement of the pelvic wall

Distant metastases (typically to the liver or lung) may or may not be a relative contraindication depending on the location and potential for curative resection. (See "Potentially resectable colorectal cancer liver metastases: Integration of surgery and chemotherapy" and "Surgical resection of pulmonary metastases: Outcomes by histology".)

●For most patients, especially those who have not received prior pelvic radiation therapy (RT), we suggest combined modality therapy rather than surgery alone (Grade 2C). (See 'Combined modality therapy' above.)

For previously unirradiated patients, we follow the same principles as guide preoperative therapy of primary locally advanced rectal adenocarcinoma. Preoperative therapy with fluoropyrimidine-based, concurrent, long-course external beam RT followed by surgery six to eight weeks later and then four to six months of adjuvant chemotherapy is preferred over surgery followed by adjuvant therapy. For patients with bulky tumors or extensive nodal disease, another option is to start with up to four months of systemic oxaliplatin-containing chemotherapy, followed by long-course chemoradiotherapy and surgery. Another option is short-course RT followed by fluoropyrimidine and oxaliplatin-based chemotherapy. (See 'Patients with no prior radiation therapy' above.).

For previously irradiated patients, pelvic reirradiation is feasible in selected patients and may permit surgical salvage and long-term survival. Patients being considered for reirradiation need to be carefully selected, and careful attention to dose and technique is required to minimize the risk of serious late effects. Sophisticated RT techniques, such as intensity-modulated RT or proton beam irradiation, are required to stay within normal tissue tolerance constraints. The dose of RT for reirradiation in this setting should be limited to 30 to 39 Gy. (See 'Previously irradiated patients' above.)

●For patients who do not receive preoperative chemotherapy, the benefit of adjuvant chemotherapy is not established. For most patients, we would consider adjuvant chemotherapy only in a patient who had not received treatment with an oxaliplatin-based chemotherapy regimen (eg, oxaliplatin plus leucovorin and short-term infusional fluorouracil [FOLFOX], oxaliplatin plus capecitabine [CAPOX]) after initial resection. (See 'Benefit of adjuvant/neoadjuvant chemotherapy' above.)

●For symptomatic patients who are not candidates for potentially curative multimodality therapy, symptom relief can often be achieved through a diverting colostomy, endoscopic laser ablation, or stent placement (for tumors not in the mid or distal rectum); palliative tumor resection is uncommonly needed. (See 'Palliation of obstructive symptoms due to locally advanced disease' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Miguel A Rodriguez-Bigas, MD, and David P Ryan, MD, who contributed to earlier versions of this topic review.