Future Directions in Adjuvant Therapy for Rectal Cancer

ABSTRACT: The US National Cancer Institute Gastrointestinal Intergroup has contributed to the development of chemotherapy and radiation regimens for the treatment of stage II and III rectal cancer. The first Intergroup trial demonstrated improvement in relapse-free and overall survival for patients who received protracted venous infusion fluorouracil (5-FU) with radiation compared to those treated with bolus 5-FU. The second trial, INT 0114, demonstrated that there was no difference among four regimens, including 5-FU with or without leucovorin or levamisole (Ergamisol) vs three drugs given as concomitant therapy. High-risk (T3, N+; T4) and low-risk (T1/2, N+; T3, N0) tumor categories were identified, revealing significant differences in 5- and 7-year survival rates. Risk of local failure was also greater in the high-risk group. Pathologic assessment of lymph nodes in the surgical specimens had an important impact on these trial results. Relapse-free and overall survival were significantly different among the lymph node-negative patients, as determined by the number of lymph nodes examined. The National Surgical Adjuvant Breast and Bowel Project trials (R-01, R-02) have shown that the addition of radiation does not affect disease-free or overall survival but improves local recurrence rates. Future trials will explore use of newer agents, including capecitabine (Xeloda), irinotecan, and oxaliplatin. In addition, prospective evaluation of laboratory correlates, including molecular markers, will be integrated into trial design. [ONCOLOGY 16(Suppl 5):45-51, 2002]

Through the years, the National Cancer Institute (NCI)Gastrointestinal Intergroup has designed a series of sequential clinical trialsto explore chemotherapy and radiation regimens for patients with stage II andIII rectal cancer. Intergroup participants have included Cancer and LeukemiaGroup B (CALGB), the Eastern Cooperative Oncology Group (ECOG), the NationalCancer Institute of Canada Clinical Trials Group (NCIC-CTG), the NationalSurgical Adjuvant Breast and Bowel Project (NSABP), the North Central CancerTreatment Group (NCCTG), the Radiation Therapy Oncology Group (RTOG), and theSouthwest Oncology Group (SWOG). Most recently, the American College of SurgeryOncology Group (ACOSOG) has joined the Intergroup network. This articlesummarizes the Gastrointestinal Intergroup continuum of rectal cancer trials,with an outline of current or planned future trials.

Improved Survival With Protracted Venous 5-FU Infusion

The first Intergroup trial of adjuvant therapy for rectal cancer, initiatedin 1986, was coordinated by NCCTG and included CALGB, ECOG, RTOG, and SWOG.[1] Atotal of 660 patients with stage II or III rectal cancer were randomly assignedto receive bolus 5-FU with or without semustine (methyl-CCNU) for two cyclesprior to radiation treatment. Protracted venous infusion 5-FU or bolus 5-FU wascombined with radiation. After radiation, two additional 5-FU cycles with orwithout semustine were administered (Table 1). Results showed significantlyimproved relapse-free survival (63% vs 53%, P =.01) among 328 patients whoreceived protracted venous infusion 5-FU with radiation compared with 332patients who received bolus 5-FU (Figure 1). Overall survival also favored theprotracted venous infusion fluorouracil regimen (70% vs 60%, P = .005). Theaddition of semustine did not improve relapse-free or overall survival (P = .33,P = .61, respectively). The regimens were well tolerated, although diarrheaoccurred more often in patients receiving protracted venous infusionfluorouracil with radiation and leukopenia was more predominant with the bolusregimen.

Identification of Tumor Risk Categories in INT 0114

Multiple analyses of results of the second Intergroup rectal cancer adjuvanttherapy trial have been published (INT 0114). This study included the samecooperative group participants as the first, plus the NCIC-CTG.[2-4] It wasdesigned to assess the role of biochemical modulation of 5-FU and that oflevamisole (Ergamisol) in postoperative rectal cancer patients, building uponcolon cancer adjuvant therapy data. Because the first Intergroup trial resultswere unavailable at the time of initiation, INT 0114 did not incorporateprotracted venous 5-FU infusion during radiation. The same "sandwich"approach was taken, incorporating two cycles of chemotherapy prior to combinedchemotherapy and radiation followed by two additional chemotherapy cycles. Thefour regimens tested included bolus 5-FU with or without leucovorin orlevamisole vs the three drugs given as concomitant therapy (see Table1).

Levamisole was not administered during radiation treatment, but 5-FU with orwithout leucovorin was. Accrual of 1,792 patients was completed in 1992, and1,695 were evaluable. Patients had pathologically confirmed stage II or IIIrectal cancer (T3/4, N0-3, M0). With a median follow-up of 7.4 years,[4]overall disease-free survival was not statistically significantly differentamong the four treatment groups; however, there were significant differences bystage. High-risk (T3, N+; T4) and low-risk (T1/2, N+; T3, N0) groups of patientswere identified, who also had significant differences in 5- and 7-year survivalrates (P < .0001). For example, 5- and 7-year survival rates were 76% vs 70%in the low-risk group, and 55% vs 45% in the high-risk group, respectively.

Results also showed a significant difference in risk of local failure in thelow- and high-risk groups (9% vs 18%, respectively; P < .0001) (Table2). Theoverall local failure rate increased over time (14% at 5 years vs 17% at 7years). Patients with T4 disease had the highest risk of local failure, a 24%local failure rate at 5 years. This rate is of concern because it issignificantly higher than reported in other studies (eg, study R-02).

Toxicities occurred significantly more frequently and were more severe infemales than in males (P < .001). Grade 3 to 5 toxicity was noted in 81% offemales and 69% of males, suggesting that females received a biologically higherdose. There was a nonsignificant trend to improved outcome in males who received5-FU and leucovorin regimens; however, disease-free survival and localrecurrence rates were not significantly different by gender. Overall survivalwas worse for males (P = .03).

Pathologic assessment of lymph nodes in the surgical specimens yieldedimportant results. While no differences were noted among lymph node-positivepatients, relapse-free and overall survival differed significantly among lymphnode-negative patients as determined by the number of lymph nodes examined.For example, 5-year survival rate was 68% for patients with  mire than 0but under  5lymph nodes analyzed, compared with 82% for patients with > 14 lymph nodesanalyzed (Table 3).[2]

Ongoing Trial of Postoperative Infusional 5-FU (S9304)

The third Intergroup trial, coordinated by SWOG (S9304), completed accrual in2000 with 1,700 patients. The treatment design was based on the previousIntergroup trials. The control arm included bolus 5-FU given before and afterradiation with protracted venous infusion 5-FU throughout radiation. Anexperimental arm incorporated protracted venous infusion 5-FU throughout thetreatment course. The third arm included 5-FU, leucovorin, and levamisole, withbolus 5-FU and leucovorin during radiation, similar to INT 114 (see Table1).Results of this trial, when available, should help define the role of infusional5-FU for rectal cancer patients in the postoperative setting.

NSABP Adjuvant Therapy Trials in RectalCancer

The NSABP has conducted three trials of adjuvant therapy for rectal cancer.In the first trial (R-01), 574 patients with stage II or III rectal cancer wererecruited from 1977 to 1986.[5] Patients were randomly assigned to receive nofurther therapy after surgery, postoperative adjuvant chemotherapy with MOF(semustine [methyl-CCNU], vincristine [Oncovin], and 5-FU), or postoperativeradiation therapy without chemotherapy (see Table1). Results demonstratedimproved disease-free survival (42% vs 30%, P = .006) and overall survival (52%vs 43%, P = .05) in the MOF arm compared with the surgery-alone arm (Figure2).Only males, however (particularly younger males), experienced this survivalbenefit. Radiation therapy was shown to reduce locoregional recurrence (25% vs16%, P = .06), but provided no advantage in disease-free or overall survival.

In another trial conducted by the NSABP (R-02) from 1987 through 1992,[6] 694stage II and III rectal cancer patients were randomly assigned to receiveadjuvant chemotherapy with or without radiation. The adjuvant chemotherapy forfemales consisted of 5-FU and leucovorin; in males it was MOF vs 5-FU andleucovorin (see Table 1). As in R-01, the addition of radiation did not improvedisease-free or overall survival, while it significantly reduced risk oflocoregional recurrence from 13% to 8% at 5 years (P = .02) (Figure3).Disease-free survival at 5 years was significantly improved for male patientswho received 5-FU and leucovorin compared to MOF (55% vs 47%, respectively; P =.009), but overall survival was not affected (65% vs 62%, respectively; P =.17).

Diarrhea was noted more often in patients treated with 5-FU and leucovorinthan in those receiving MOF, while leukopenia and thrombocytopenia were seen inmore MOF than 5-FU/leucovorin patients. Grade 3 toxicities occurred morefrequently in women than in men (34% vs 22%, respectively). The rate of grade 3toxicities was comparable (approximately 30%) in patients receiving chemotherapyalone or the combination of radiation and 5-FU/leucovorin.

The NSABP attempted to compare preoperative chemoradiation to postoperativecombined-modality therapy in the R-03 trial (see Table1). A similar trial wasattempted by the Intergroup (R-9401). Preoperative chemoradiation is ofconsiderable interest because of the potential to down-stage patients,increasing the possibility of sphincter-sparing surgery. This approach may alsoreduce chronic toxicities because irradiated tissue can be removed duringsurgery. On the other hand, postoperative therapy provides the opportunity forpathologic staging, thereby avoiding radiation therapy in patients with stage Idisease. Accrual was not completed in either of these trials.

Preliminary results of R-03 were reported at the annual American Society ofClinical Oncology meeting in 2001.[7] A total of 267 patients were enrolled from1993 to 1999. At 1-year of follow-up, sphincter-saving surgery was possible inmore patients who received preoperative 5-FU, leucovorin, and radiation than inthose who received postoperative therapy (44% vs 34%, respectively). Among thepreoperative therapy group, 23% had a complete clinical response and 16.2% had acomplete pathologic response. More of the preoperatively treated patients hadgrade 4/5 toxicities (34% vs 22%, respectively), although disease-free survivalwas comparable in the preoperative (83%) and postoperative (78%) therapy arms.

It is doubtful that a large, definitive, randomized trial comparingpreoperative chemoradiation to postoperative combined therapy will be achievable—atleast not in the United States. A recent meta-analysis of preoperativechemoradiation vs surgery alone demonstrated a significant reduction in overallmortality favoring preoperative radiation (14 randomized trials, P = .03).[8]The meta-analysis also reviewed cancer-related mortality and local recurrence in11 of the randomized trials, which revealed an advantage for both parametersfavoring preoperative therapy (P < .001 for both parameters).

Other Ongoing and Planned Trials

The Medical Research Council (MRC) CR07 trial, being conducted in the UnitedKingdom, will attempt to explore further preoperative radiation vs selectivepostoperative chemo/radiotherapy. Patients will be randomly assigned to receive25 Gy of radiation therapy (in five fractions) followed by surgery with localpathology assessment, and adjuvant therapy per physician discretion. The secondstudy arm will also include pathologic assessment for postoperative patients.Select patients in the postoperative chemo/radiotherapy arm considered to be athigh risk of local recurrence following surgery will receive 45 Gy of radiationtherapy (in five fractions) plus a chemotherapy regimen chosen by theirphysician. The target accrual is 1,800 patients, and the study end point islocal recurrence.

German researchers are planning a trial of 800 stage II and III rectal cancerpatients who will receive preoperative treatment with 50.4 Gy of radiation pluscontinuous infusion 5-FU administered during the first and fifth weeks ofradiation, or the same regimen postoperatively. In addition, both regimens willincorporate four cycles of 5-day 5-FU administered in the postoperative setting.

Future and ongoing trials in the United States will evaluate the contributionof new agents and incorporate tissue acquisition for correlative laboratoryprojects. For example, ECOG is conducting a phase I trial for patients with T3or T4 rectal carcinoma. Treatment consists of 50.4 Gy of radiation therapy withprotracted venous infusion 5-FU and oxaliplatin, and postoperative weeklyhigh-dose leucovorin plus 5-FU.

Another trial being planned by the entire US Gastrointestinal Intergroup,including NSABP, allows the physician and patient to choose between preoperativeor postoperative chemoradiation. Chemotherapy administered during radiation mayinclude capecitabine, bolus 5-FU with or without leucovorin or continuousinfusion 5-FU. Randomization will compare a standard regimen of high-doseleucovorin (500 mg/m²) and 5-FU (500 mg/m²) administered weekly for 6 weeks fora total of four cycles, vs irinotecan, 5-FU, and leucovorin using a Europeaninfusion regimen (Figure 4). This every-other-week infusion schedule was chosenin part because of toxicity concerns with weekly bolus irinotecan and asuggestion that toxicity may be reduced with an infusion program.[9,10]

Efforts to develop oral fluoropyrimidines as potential alternatives toprotracted venous infusion 5-FU are continuing. Capecitabine (Xeloda), a prodrugconverted to 5-FU by a three-enzyme pathway, is the only commercially availableoral drug for the treatment of advanced colorectal cancer. In two largerandomized trials for patients with advanced colorectal cancer, oralcapecitabine produced a higher response rate than the 5-day bolus regimen of5-FU/leucovorin (eg, 24.8% vs 15.5%, respectively; P = .005) with a morefavorable toxicity profile.[11,12] Furthermore, time to disease progression andsurvival rates were similar with the two regimens. In addition, capecitabine isviewed as a potential radiosensitizer because it results in selectiveconcentration of 5-FU in tumor tissue, and 5-FU is a known radiation sensitizer.Phase I data from Germany suggest using a phase II capecitabine dose of 825mg/m² bid during radiation therapy for rectal cancer (50.4 Gy).[13]

The NSABP has designed the R-04 trial, which is a preoperative chemoradiationtrial comparing radiation (45 Gy) plus oral capecitabine at 825 mg/m² bidthroughout radiation vs protracted venous infusion 5-FU at 225 mg/m² (Figure5).End points include toxicity assessment and rates of pathologic complete responseand successful sphincter-saving surgery. A second randomization will includeadministration of epoetin alfa in conjunction with preoperative radiation andchemotherapy to determine whether it improves hemoglobin levels, quality oflife, and local tumor control. Tumor tissue samples will be available toevaluate gene expression profiles.

The RTOG is exploring the use of irinotecan and continuous infusion 5-FU withpreoperative radiation in patients with T3 or T4 rectal cancer (R-0012). Thisrandomized phase II trial will evaluate protracted venous infusion 5-FU (250mg/m²/d) with radiation vs irinotecan (50 mg/m² IV every week for four courses)plus protracted venous infusion 5-FU (225 mg/m²/d Monday through Friday)combined with radiation.

Many trials will continue to explore biologic correlates for predictors ofsurvival and response and for potential therapeutic intervention. For example,in the R-01 study, thymidylate synthase (TS) expression was evaluated byimmunohistochemistry: patients with low TS staining intensity had improveddisease-free (49% vs 27%, P < .01, respectively) and overall survival (60% vs40%, P < .01, respectively).[14] In addition, patients with high TSexpression were more likely to benefit from chemotherapy; in high-TS patients,54% of those treated with chemotherapy survived compared with 31% treated withsurgery alone (P < .01). Prospective analyses of the predictive power of TSand other markers (eg, dihydropyrimidine dehydrogenase, thymidine phosphorylase)are planned for both Intergroup and NSABP trials.

Recent molecular data from trials assessing adjuvant therapy for colon cancersuggest that patients with retention of the 18q allele and those withmicrosatellite instability with TGF-beta1 RII mutation have significant improvementin 5-year survival when treated with 5-FU-based adjuvant therapy.[15] Trialsof adjuvant therapy for both colon and rectal cancer will now prospectivelyanalyze these and other molecular markers to clarify these provocativeobservations.

References:

1. O’Connell MJ, Martenson JA, Wieand HS, et al: Improving adjuvant therapyfor rectal cancer by combining protracted-infusion fluorouracil with radiationtherapy after curative surgery. N Engl J Med 331:502-507, 1994.

2. Tepper JE, O’Connell MJ, Niedzwiecki D, et al: Impact of number of nodesretrieved on outcome in patients with rectal cancer. J Clin Oncol 19:157-163,2001.

3. Tepper JE, O’Connell MJ, Petroni GR, et al: Adjuvant postoperativefluorouracil-modulated chemotherapy combined with pelvic radiation therapy forrectal cancer: Initial results of Intergroup 0114. J Clin Oncol 15:2030-2039,1997.

4. Tepper JE, O’Connell MJ, Niedzwiecki D, et al: Adjuvant therapy inrectal cancer: Analysis of stage, sex, and local control—Final report ofIntergroup 0114. J Clin Oncol 20:1744-1750, 2002.

5. Fisher B, Wolmark N, Rockette H, et al: Postoperative adjuvantchemotherapy or radiation therapy for rectal cancer: Results from NSABP ProtocolR-01. J Natl Cancer Inst 80:21-29, 1988.

6. Wolmark N, Wieand HS, Hyams DM, et al: Randomized trial of postoperativeadjuvant chemotherapy with or without radiotherapy for carcinoma of the rectum:National Surgical Adjuvant Breast and Bowel Project Protocol R-02. J Natl CancerInst 92:388-396, 2000.

7. Roh MS, Petrelli N, Wieand S, et al: Phase III randomized trial ofpreoperative versus postoperative multimodality therapy in patients withcarcinoma of the rectum (NSABP R-03) (abstract 490). Proc Am Soc Clin Oncol20:123a, 2001.

8. Camma C, Giunta M, Fiorica F, et al: Preoperative radiotherapy forresectable rectal cancer: A meta-analysis. JAMA 284:1008-1015, 2000.

9. Rothenberg ML, Meropol NJ, Poplin EA, et al: Mortality associated withirinotecan plus bolus fluorouracil/leucovorin: Summary findings of anindependent panel. J Clin Oncol 19:3801-3807, 2001.

10. Douillard JY, Cunningham D, Roth AD, et al: Irinotecan combined withfluorouracil compared with fluorouracil alone as first-line treatment formetastatic colorectal cancer: A multicentre randomised trial. Lancet355:1041-1047, 2000.

11. Hoff PM, Ansari R, Batist G, et al: Comparison of oral capecitabineversus intravenous fluorouracil plus leucovorin as first-line treatment in 605patients with metastatic colorectal cancer: Results of a randomized phase IIIstudy. J Clin Oncol 19:2282-2292, 2001.

12. Van Cutsem E, Twelves C, Cassidy J, et al: Oral capecitabine comparedwith intravenous fluorouracil plus leucovorin in patients with metastaticcolorectal cancer: Results of a large phase III study. J Clin Oncol19:4097-4106, 2001.

13. Dunst J, Reese T, Frings S: Phase I study of capecitabine combined withstandard radiotherapy in patients with rectal cancer (abstract 995). Proc Am SocClin Oncol 19:256a, 2001.

14. Johnston PG, Fisher ER, Rockette HE, et al: The role of thymidylatesynthase expression in prognosis and outcome of adjuvant chemotherapy inpatients with rectal cancer. J Clin Oncol 12:2640-2647, 1994.

15. Watanabe T, Wu T, Catalano PJ, et al: Molecular predictors of survivalafter adjuvant chemotherapy for colon cancer. N Engl J Med 344:1196-1206, 2001.

16. Lee-Wade E, Benson III AB: Colorectal cancer: Chemotherapy, in KassirerJP, Greene HL III (eds): Current Therapy in Adult Medicine, 4th ed.Philadelphia, Mosby Year Book Inc, 1997.