Over the past decade, patients with locally advanced rectal cancer at The University of Texas M. D. Anderson Cancer Center have been managed with preoperative chemoradiation. Patients achieving a complete clinical response to preoperative chemoradiation have had better pelvic tumor control, sphincter preservation, and overall survival than those with gross residual disease. Some patients achieving a complete clinical response have even had rectal-preserving surgery (full-thickness local excision).
ABSTRACT: Over the past decade, patients with locally advanced rectal cancer at The University of Texas M. D. Anderson Cancer Center have been managed with preoperative chemoradiation. Patients achieving a complete clinical response to preoperative chemoradiation have had better pelvic tumor control, sphincter preservation, and overall survival than those with gross residual disease. Some patients achieving a complete clinical response have even had rectal-preserving surgery (full-thickness local excision). These results emphasize the importance of maximizing tumor response. Further improvement in response and survival could be achieved by using novel chemotherapeutic agents or through tumor-selective molecular targeting strategies that enhance the effects of chemotherapy, radiotherapy, or both. Irinotecan (CPT-11, Camptosar) is a novel chemotherapy agent being evaluated clinically as a radiosensitizing agent in rectal cancer. Inhibition of several molecular targets-such as epidermal growth factor receptor, ras oncogene activation, the cyclooxygenase-2 (COX-2) enzyme, and neoangiogenesis-appears to be tumor-selective in preclinical models. COX-2 expression has been shown to enhance cytotoxic therapy in preclinical models. In vitro and in vivo studies show that selective COX-2 inhibition enhances the effects of radiotherapy as well as chemotherapy. COX-2 is also markedly upregulated in human colorectal cancer and appears to be associated with adverse patient prognosis. Thus, integration of molecular targeting, such as COX-2 selective inhibition with existing chemoradiation approaches, may provide selective tumor radiosensitization and chemosensitization, resulting in improved pelvic control, sphincter preservation, and overall survival. [ONCOLOGY 16(Suppl 5):39-44, 2002]
Preoperative chemoradiation has become a commonapproach to treat stage II and III rectal cancer at many institutions, includingThe University of Texas M. D. Anderson Cancer Center (MDACC) in Houston.[1-4] Astudy from Uppsala, Sweden, has shown that preoperative radiation alone improvespelvic control as compared with postoperative radiation alone. AnotherSwedish trial in patients with rectal cancer demonstrated improved pelviccontrol and overall survival using preoperative radiotherapy compared withsurgery alone. These studies, however, used preoperative radiotherapy withoutconcurrent chemotherapy.
Although a randomized trial has not been conducted to compare the efficacy ofpreoperative chemoradiation to that of radiotherapy alone, concurrentchemoradiation is commonly used in the United States. This practice is based onextrapolation from trials in the postoperative setting[7,8] showing thesuperiority of combined chemotherapy and radiotherapy over radiotherapy alone.In this respect, the use of a protracted venous infusion of fluorouracil (5-FU)has become the standard chemotherapy, based on data from a randomized trialconducted by the Gastrointestinal Intergroup.
Patients who have had a complete clinical response to preoperativechemoradiation have had better pelvic tumor control, sphincter preservation, andoverall survival than those who have had gross residual disease. Some patientsachieving a clinical complete response have even had rectal preserving surgery(full-thickness local excision). These results emphasize the importance ofmaximizing tumor response. The improvement in overall survival could be due tohigher pelvic control rates, or the response may indicate increased sensitivityof distant micrometasteses to chemotherapy. Any increase in the response oflocal disease to chemoradiation or distant micrometasteses to chemotherapy wouldlogically further improve outcome.
Use of chemotherapeutic agents that may be more effective than 5-FU asradiosensitizers or as systemic agents may also lead to further improvements inresponse and survival rates. Alternatively, molecular targeting could be used toselectively enhance the efficacy of chemoradiotherapy regimens. Irinotecan(CPT-11, Camptosar) is among the newer chemotherapeutic agents being explored incombination with radiotherapy. Phase I studies have determined the maximumtolerated dose of irinotecan when used with preoperative chemoradiation inrectal cancer patients. The Radiation Therapy Oncology Group (RTOG) iscurrently evaluating irinotecan as a radiosensitizer in rectal cancer.
Because the primary limitation of chemoradiation is toxicity to normaltissues, an improvement in therapeutic ratio could also be achieved byincorporating tumor-selective agents. A number of processessuch as epidermalgrowth factor receptor signaling, ras oncogene activation, the cyclooxygenase-2(COX-2) enzyme, and angiogenesisappear to be tumor-selective in preclinicalmodels. Treatment with inhibitors of these molecular processes has showntherapeutic efficacy in several preclinical tumor models, particularly whencombined with cytotoxic agents.
Among molecular target therapeutic strategies, counteracting COX-2 enzymeactivity seems promising, and significant preclinical and clinical testing ofthis approach has been initiated. COX-2 overexpression occurs in various typesof human cancer and has been shown to correlate with poor prognosis.Study results have shown that treatment of tumor-bearing mice with selectiveCOX-2 inhibitors enhances tumor response to radiation and chemotherapy.Cyclooxygenase inhibitors have been shown to prevent colon cancer[16,17] and toreduce the number of polyps in patients with familial adenomatous polyposis.Therefore, COX-2 inhibition has the potential to improve outcome for rectalcancer patients treated with preoperative chemoradiation.
Over the past decade at MDACC, patients with rectal cancer have been treatedwith a preoperative radiotherapy dose of 45 Gy in 25 fractions prescribed to the95% isodose line, with concomitant protracted venous infusion 5-FU at 300mg/m2/d, Monday through Saturday morning. Since 1997, patients have been treatedin an ongoing prospective phase II trial that includes a concomitant boosttechnique. With this technique, the same radiation dose is delivered to thetumor and pelvic lymphatics (45 Gy) over 5 weeks, with an additional 1.5-Gyboost as a second daily treatment separated from the first by 6 hours during thefifth treatment week. The total dose to the primary tumor is 52.5 Gy prescribedto the 95% isodose line. The boost is given to the gross tumor volume using a 3Dconformal technique, which minimizes normal tissue irradiation. Mesorectalexcision is carried out 4 to 8 weeks later. All patients receive four cycles ofadjuvant 5-FU and leucovorin after surgery.
Between 1990 and 2000, 390 stage II and III rectal cancer patients weretreated with preoperative pelvic chemoradiation followed by surgery. Presence ofmicroscopic or no residual disease was shown to be a favorable predictor ofpelvic control on univariate analysis (P = .02, Figure1). A Cox regressionmultivariate analysis showed a trend toward improved survival (P = .08,unpublished observations). The only independently significant prognostic factorfor freedom from pelvic recurrence was clinical T stage. Actuarial freedom frompelvic recurrence at 4 years was 90% for 354 patients with T3 disease vs 79% for36 patients with T4 disease. At MDACC, clinical T4 disease is defined asobjective evidence of adjacent organ involvement on computed tomography (CT)scan, physical examination, or endoscopic ultrasound.
While the concept is not universally accepted, several investigators havereported that significant numbers of patients who were initially thought torequire abdominoperineal resection were able to undergo sphincter-preservingprocedures following preoperative chemoradiation. This phenomenon has beendemonstrated prospectively by preliminary results of the ongoing NationalSurgical Adjuvant Breast and Bowel Project (NSABP) R-03 trial, in which patientswere randomly assigned to receive preoperative or postoperative chemoradiation.Surgeons entering patients in the study were encouraged to decide prospectivelywhich operation would be necessary. A sphincter-preserving procedure waspossible in 23% of those patients initially felt to require abdominoperinealresection.
While studies have suggested that response to preoperative chemoradiation mayimpact the rate of sphincter-preserving surgery,[2,22] those analyses were notoriginally designed to answer that question. At MDACC, the sphincterpreservation rate in patients with low rectal cancer following preoperativechemoradiation has increased with time and is currently greater than 60% fortumors < 6 cm or less from the anal verge. A logistic regression analysis ofclinical, treatment-related, and pathologic factors among 238 patients withtumors < 6 cm from the anal verge indicated that complete clinical responseto preoperative chemoradiation is an independent predictor of sphincterpreservation.
Clinical complete response rate was 46%, and tumors < 3 cm or less from the analverge that responded completely to preoperative chemoradiation had the moststriking difference in sphincter-preserving surgery (44% vs 23%, P = .01)compared to those who did not respond completely. Importantly, the pelvicrecurrence rate among patients undergoing sphincter-preserving surgery has beenless than 10% since 1993, and no increase in pelvic recurrence risk has beenobserved with the use of sphincter-preserving procedures in the low rectum(unpublished observations). Thus, increasing the response rate could lead toincreased rates of sphincter preservation or even rectal preservation.
Of 362 patients with T3 rectal cancer treated at MDACC with preoperativechemoradiation and surgery over the past decade, 17 patients with localizeddisease have been selected for local excision due to patient refusal ofabdominoperineal resection, medical comorbidity, physician preference after acomplete clinical response, or unknown reasons. Local excision was performedafter completion of chemoradiation. With a median follow-up of 34 months (range:6 to 96 months), only 1 of 17 patients has had intrapelvic recurrence andanother has had disease recur in an inguinal lymph node. Two patients have diedof metastatic rectal cancer and one has died of unknown causes.
The 5-year actuarial overall survival at 5 years was 67%. Twelve of 17 (71%)patients have achieved pathologic complete response, microscopic residualdisease was evident in 4/17 (24%), and gross residual disease was present in onepatient. This indicates that the response to preoperative chemoradiation mayhave played a role in the decision to proceed with local excision rather thanabdominoperineal resection in some patients. Furthermore, it appears thatpreoperative chemoradiation followed by local excision provides effective localcontrol in highly selected cases. This strategy should be studied prospectivelyin patients with low rectal cancer who respond to chemoradiation.
Previously, we reported that response to preoperative chemoradiation (definedas evidence of microscopic or no residual disease) is a positive predictor ofimproved overall survival (P = .07). At the time of that analysis, thefollow-up duration and patient numbers were insufficient to demonstratestatistical significance. We recently reported results involving a largerpatient cohort, which showed that response (defined as evidence for T-stagedown-staging without nodal positivity) is a positive predictor ofdisease-free and overall survival. The presence of microscopic or noresidual disease also positively predicted survival on univariate analysis inthis updated experience (P = .002, Figure2). A Cox regression multivariateanalysis confirmed this finding as well as the negative influence of poordifferentiation on overall survival (unpublished observations).
In in vivo preclinical studies conducted at MDACC, irinotecan hadsignificant, dose-dependent radiosensitizing effects. Dose modification factorsas high as 3.7 have been demonstrated in human tumor xenografts. Clinically,irinotecan has been well tolerated when administered with concurrent pelvicradiotherapy. In a recent phase I study evaluating continuous infusion 5-FU andradiotherapy (45 Gy to the pelvis) with dose escalation of irinotecan, therecommended dose of weekly irinotecan was 50 mg/m². Grade 3 or 4 diarrhea orintravenous line infection occurred in 3 of 11 patients treated at thatdose. Furthermore, an ongoing study at MDACC in advanced gastric cancerpatients indicates that weekly irinotecan doses of at least 60 mg/m² are welltolerated with whole-stomach irradiation (unpublished data).
The importance of irinotecan for metastatic disease is illustrated by twoimportant well-designed and conducted randomized trials that demonstrated amedian survival benefit when compared to best supportive care and afterfailure on 5-FU. One of these studies showed improved quality of life aswell. The addition of weekly irinotecan to first-line 5-FU and leucovorintherapy has been shown to improve progression-free survival and overall survivalof metastatic colorectal cancer patients. The combination of irinotecan,5-FU, and leucovorin has become the standard of care in patients with metastaticcolorectal cancer.
Prostaglandins are arachidonic acid metabolites that have a variety ofbiologic capabilities. Prostaglandins have been implicated in the promotion,development, and growth of malignant tumors,[30-33] and have also been shown toinfluence the response of normal tissues to radiation. Two enzymes mediateprostaglandin production: COX-1, which is ubiquitous and constitutivelyexpressed, and COX-2, which is inducibly expressed by inflammatorystimuli.[33,34] Increasing evidence indicates that agents that inhibit COX-2 caninhibit cancer development in both animals and humans, reduce the size ofestablished tumors,[32,33,35] and enhance cytotoxic efficacy of anticanceragents.
Studies conducted at MDACC have demonstrated that the nonsteroidalanti-inflammatory drug (NSAID) indomethacin prolongs tumor growth delay andincreases tumor cure rate in mice treated with radiotherapy.[35-37] StandardNSAIDs have limited clinical utility, however, because they inhibit both COX-1and COX-2. COX-1 inhibition induces nonselective toxicity and a consequentreduced ability to achieve therapeutic gain without severe toxicity to normaltissues. The recently developed selective COX-2 inhibitors have the potential toovercome this limitation.
Recent experimental evidence shows that COX-2 inhibition can preventcarcinogenesis in experimental animals. Studies at MDACC demonstratedenhancement of radiotherapy effects with the addition of a COX-2-specificinhibitor. Using a murine sarcoma model, the enhancement ratio was 3.64 using agrowth delay assay and 1.77 using a tumor cure assay (TCD50) (Figure3). Thecorresponding normal tissue enhancement was studied in a jejunal crypt assay andfound to be minimal, if any (enhancement factor of 1.03). Thus, this novelapplication of a COX-2 inhibitor resulted in rather selective radioenhancement.
Attention has only recently been focused on understanding the mechanisms ofthis sensitization and evaluating the potential interaction with commonly usedradiosensitizers, such as 5-FU and novel radiosensitizers like irinotecan. COX-2inhibition combined with many systemic agents, including irinotecan, had asupra-additive effect in vitro. Masferrer (personal communication, July2001) has discussed similar findings. Thus far, no clinical studies have beenreported investigating the effect of selective COX-2 inhibition when combinedwith radiotherapy, chemotherapy, or chemoradiotherapy.
Patients with rectal cancer who respond to preoperative chemoradiationtreatment have better pelvic control, sphincter preservation, and overallsurvival than those who do not respond. Use of local excision instead ofmesorectal excision appears to be safe in selected patients who respond topreoperative chemoradiation. However, additional improvements in the treatmentof rectal cancer are needed, and emerging strategies include the use ofirinotecan and COX-2 inhibitors. The addition of irinotecan to 5-FU andleucovorin has improved survival in patients with localized resected coloncancer, and irinotecan is being evaluated prospectively for its role as aradiosensitizer.
Incorporation of molecular targeting into chemoradiotherapy treatments isanother rapidly emerging strategy, and COX-2 represents one of the promisingtargets. Inhibitors of COX-2 have been found to enhance the effects ofradiotherapy, irinotecan, 5-FU, and other cytotoxic agents when tested in vitroand in vivo. The integration of COX-2 inhibition with existing chemoradiationprograms may improve outcomes even further.
1. Minsky BD, Coia L, Haller D, et al: Treatment systems guidelines forprimary rectal cancer from the 1996 Patterns of Care Study. Int J Radiat OncolBiol Phys 41(1):21-27, 1998.
2. Vauthey JN, Marsh RW, Zlotecki RA, et al: Recent advances in the treatmentand outcome of locally advanced rectal cancer. Ann Surg 229(5):745-754, 1999.
3. Rich TA, Skibber JM, Ajani JA, et al: Preoperative infusionalchemoradiation therapy for stage T3 rectal cancer. Int J Radiat Oncol Biol Phys32(4):1025-1029, 1995.
4. Janjan NA, Khoo VS, Abbruzzese J, et al: Tumor downstaging and sphincterpreservation with preoperative chemoradiation in locally advanced rectal cancer:The M. D. Anderson Cancer Center experience. Int J Radiat Oncol Biol Phys44(5):1027-1038, 1999.
5. Frykholm G, Glimelius B, Pahlman L: Preoperative or postoperativeirradiation in adenocarcinoma of the rectum: Final treatment results of arandomized trial and an evaluation of late secondary effects. Dis Colon Rectum36(6):564-572, 1993.
6. Improved survival with preoperative radiotherapy in resectable rectalcancer. Swedish Rectal Cancer Trial [see comments] N Engl J Med 336(14):980-987,1997 (published erratum appears in N Engl J Med 336:1539, 1997).
7. Krook JE, Moertel CG, Gunderson LL, et al: Effective surgical adjuvanttherapy for high-risk rectal carcinoma [see comments]. N Engl J Med324(11):709-715, 1991.
8. Prolongation of the disease-free interval in surgically treated rectalcarcinoma. Gastrointestinal Tumor Study Group. N Engl J Med 312(23):1465-1472,1985.
9. O’Connell M, 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(8):502-507, 1994.
10. Mitchell E, Anne P, Fry R, et al: Combined modality therapy of locallyadvanced or recurrent adenocarcinoma of the rectum: Report of a phase I trial ofchemotherapy with CPT-11, 5-FU, and concomitant irradiation (abstract 519). ProcAm Soc Clin Oncol 20:131a, 2001
11. Mason K, Komaki R, Cox J, et al: Biology-based combined modalityradiotherapy: Workshop report. Int J Radiat Oncol Biol Phys 50(4):1079-1089,2001.
12. Masferrer JL, Leahy KM, Koki AT, et al: Antiangiogenic and antitumoractivities of cyclooxygenase-2 inhibitors. Cancer Res 60(5):1306-1311, 2000.
13. Sheehan KM, Sheahan K, O’Donoghue DP, et al: The relationship betweencyclooxygenase-2 expression and colorectal cancer. JAMA 282(13):1254-1257, 1999.
14. Milas L, Kishi K, Hunter N, et al: Enhancement of tumor response togamma-radiation by an inhibitor of cyclooxygenase-2 enzyme. J Natl Cancer Inst91(17):1501-1504, 1999.
15. Hida T, Kozaki K, Muramatsu H, et al: Cyclooxygenase-2 inhibitor inducesapoptosis and enhances cytotoxicity of various anticancer agents in non-smallcell lung cancer cell lines. Clin Cancer Res 6(5):2006-2011, 2000.
16. Thun MJ, Namboodiri MM, Heath CW, Jr: Aspirin use and reduced risk offatal colon cancer. N Engl J Med 325(23):1593-1596, 1991.
17. Giovannucci E, Egan KM, Hunter DJ, et al: Aspirin and the risk ofcolorectal cancer in women. N Engl J Med 333(10):609-614, 1995.
18. Steinbach G, Lynch PM, Phillips RK, et al: The effect of celecoxib, acyclooxygenase-2 inhibitor, in familial adenomatous polyposis. N Engl J Med342(26):1946-1952, 2000.
19. Sanfilippo N, Crane C, Skibber J, et al: T4 rectal cancer treated withpreoperative chemoradiation to the posterior pelvis followed by multivisceralresection: patterns of failure and limitations of treatment. Int J Radiat OncolBiol Phys 51(1), 2001.
20. Minsky BD: Sphincter preservation in rectal cancerContinued evidenceof success. Int J Radiat Oncol Biol Phys 46(2):267-268, 2000.
21. Hyams DM, Mamounas EP, Petrelli N, et al: A clinical trial to evaluatethe worth of preoperative multimodality therapy in patients with operablecarcinoma of the rectum: A progress report of National Surgical Breast and BowelProject Protocol R-03. Dis Colon Rectum 40(2):131-139, 1997.
22. Janjan NA, Crane CH, Feig BW, et al: Prospective trial of preoperativeconcomitant boost radiotherapy with continuous infusion 5-fluorouracil forlocally advanced rectal cancer. Int J Radiat Oncol Biol Phys 47(3):713-718,2000.
23. Crane C, Skibber J, Brown T, et al: An evaluation of clinical prognosticfactors and the concomitant boost technique on sphincter preservation inlow-lying rectal cancer treated with preoperative chemoradiation. 2ndInternational Symposium, Sphincter Saving Treatment in Rectal Cancer. Lyon,France, 2001.
24. Janjan NA, Abbruzzese J, Pazdur R, et al: Prognostic implications ofresponse to preoperative infusional chemoradiation in locally advanced rectalcancer. Radiother Oncol 51(2):153-160, 1999.
25. Janjan NA, Crane C, Feig BW, et al: Improved overall survival amongresponders to preoperative chemoradiation for locally advanced rectal cancer. AmJ Clin Oncol 24(2):107-112, 2001.
26. Kirichenko AV, Rich TA, Newman RA, et al: Potentiation of murine MCa-4carcinoma radioresponse by 9-amino-20(S)-camptothecin. Cancer Res57(10):1929-1933, 1997.
27. Cunningham D, Glimelius B: A phase III study of irinotecan (CPT-11)versus best supportive care in patients with metastatic colorectal cancer whohave failed 5-fluorouracil therapy. V301 Study Group. Semin Oncol 26(1 suppl5):6-12, 1999.
28. Van Cutsem E, Blijham GH: Irinotecan versus infusional 5-fluorouracil: Aphase III study in metastatic colorectal cancer following failure on first-line5-fluorouracil. V302 Study Group. Semin Oncol 26(1 suppl 5):13-20, 1999.
29. Saltz LB, Cox JV, Blanke C, et al: Irinotecan plus fluorouracil andleucovorin for metastatic colorectal cancer. Irinotecan Study Group [seecomments]. N Engl J Med 343(13):905-914, 2000.
30. Brunda MJ, Herberman RB, Holden HT: Inhibition of murine natural killercell activity by prostaglandins. J Immunol 124(6):2682-2687, 1980.
31. Honn KV, Bockman RS, Marnett LJ: Prostaglandins and cancer: A review oftumor initiation through tumor metastasis. Prostaglandins 21(5):833-864, 1981.
32. Furuta Y, Hall ER, Sanduja S, et al: Prostaglandin production by murinetumors as a predictor for therapeutic response to indomethacin. Cancer Res48(11):3002-3007, 1988.
33. Taketo MM: Cyclooxygenase-2 inhibitors in tumorigenesis (part II). J NatlCancer Inst 90(21):1609-1620, 1998.
34. Masferrer JL, Isakson PC, Seibert K: Cyclooxygenase-2 inhibitors: A newclass of anti-inflammatory agents that spare the gastrointestinal tract.Gastroenterol Clin North Am 25(2):363-372, 1996.
35. Milas L, Hanson WR: Eicosanoids and radiation. Eur J Cancer31A(10):1580-1585, 1995.
36. Milas L, Furuta Y, Hunter N, et al: Dependence of indomethacin-inducedpotentiation of murine tumor radioresponse on tumor host immunocompetence.Cancer Res 50(15):4473-4477, 1990.
37. Furuta Y, Hunter N, Barkley T, Jr, et al: Increase in radioresponse ofmurine tumors by treatment with indomethacin. Cancer Res 48(11):3008-3013, 1988.
38. Kishi K, Peterson S, Peterson C, et al: Preferential enhancement of tumorradioresponse by a COX-2 inhibitor. Cancer Res 60(5):1326-1331, 2000.