Radiation Oncology/Rectum/Adjuvant RT

Adjuvant Therapy for Rectal Cancer

Adjuvant RT Overview

• Local recurrence after surgery alone for Dukes T3-T4 or N+ is >50%, and can be symptomatically devastating
• Most adjuvant trials were done in patients with Dukes B and C (T3-4N0 or N+)
• Outcomes with adjuvant chemotherapy alone have been disappointing
• 8 prospective randomized trials evaluated surgery alone vs. post-op RT. RT alone improves local control, but not survival
• Combined RT + 5-FU improved local control, distant control, as well as survival in two randomized trials
• RT plus continued infusion of 5-FU improves survival over bolus 5-FU. Other concurrent drugs have so far not shown additional benefit over C.I. 5-FU
• For postop T3-4 or N+: Adjuvant chemotherapy (5-FU/LV), followed by continuous infusion 5-FU + RT, then additional 5-FU/LV

Surgery +/- Adjuvant RT

• Post-op RT alone improves local control, and in one trial increases time-to-local failure
• There is no impact on distant mets, disease-free survival, or overall survival
• At this time, RT alone is considered inadequate. It should be given concurrently with chemotherapy

• Meta-analysis, 2001 PMID 11684209 -- "Adjuvant radiotherapy for rectal cancer: a systematic overview of 8,507 patients from 22 randomised trials." (Colorectal Cancer Collaborative Group, Lancet. 2001 Oct 20;358(9290):1291-304.). Note: following data abstracted for post-op RT only (8 trials):
  • 5-year OS: observation 58.6% vs. RT 57.5% (NS). Annual death rate 4.6% lower with RT
  • 5-year risk of any recurrence: observation 54% vs. RT 50% (NS)
  • 5-year risk of isolated local recurrence: observation 23% vs. RT 15% (SS). This was driven by 2 trials (UK MRC 3 and NSABP R-01)
  • Few recurrences >5 years out.
  • No subgroup benefit

• MRC 3 (1984-1989) PMID 8961990 -- "Randomised trial of surgery alone versus surgery followed by radiotherapy for mobile cancer of the rectum. Medical Research Council Rectal Cancer Working Party." (No Authors Listed, Lancet. 1996 Dec 14;348(9042):1610-4.)
  • Randomized. 469 patients with Dukes B and C (T3-4N0, N+) treated with 1) surgery alone vs. 2) surgery followed 4-6 weeks later by RT 40/20. Minimum F/U 5 years
  • 5-year OS: 46% vs. 52% (NS), trend toward RT arm (HR 0.84)
  • Local recurrence: 34% surgery alone vs. 21% RT (SS). No difference in distant mets.
  • Side effects: well tolerated, late events rare

• ANZ 8202 (1982-1985) PMID 1907128 -- "Adjuvant post-operative radiotherapy in rectal cancer: results from the ANZ Bowel Cancer Trial (Protocol 8202)." (Mameghan H, Australas Radiol. 1991 Feb;35(1):61-5.)
  • Randomized. 70 patients randomized; trial terminated early due to slow accrual. 33 patients. Resection +/- RT 45 Gy. In RT group 13/36 patients did not receive RT. Average F/U 4.3 years
  • OS: 33% observation vs. 18% RT (NS). No difference in time-to-recurrence
  • Side effects: complications comparable. Grade 3-4 diarrhea: 1/21 RT patients
  • Critique: early termination, very small numbers, high noncompliance

• Rotterdam B PMID 2004322 -- "Postoperative radiation therapy for rectal cancer. An interim analysis of a prospective, randomized multicenter trial in The Netherlands." (Treurniet-Donker AD, Cancer. 1991 Apr 15;67(8):2042-8.)
  • Randomized. 172 patients. Resection +/- RT 50 Gy
  • LR lower but NS. No impact on disease-free or OS

• EORTC 40811 (1981-?) - Information from PMID 11684209
  • Randomized. 172 patients. Surgery +/- RT 46/23
  • OS: 40% vs. 37% (NS)

• Denmark (1979-?)
  • Randomized. 494/861 patients with Dukes B and C. Surgery with observation vs. post-op RT 50 Gy
  • 1992 PMID 1733432 -- "Time to loco-regional recurrence after resection of Dukes' B and C colorectal cancer with or without adjuvant postoperative radiotherapy. A multivariate regression analysis." (Bentzen SM, Br J Cancer. 1992 Jan;65(1):102-7.)
    • No OS benefit, no improvement in 5-year LR rate with RT
    • In high-risk patients, RT benefit in LR, but not OS
    • High risk: PNI, LVI, tumor <10cm from anal verge, large tumor size, necessity to resect neighbour organs
  • 1988 PMID 3166910 -- "A regression analysis of prognostic factors after resection of Dukes' B and C carcinoma of the rectum and rectosigmoid. Does post-operative radiotherapy change the prognosis?" (Bentzen SM, Br J Cancer. 1988 Aug;58(2):195-201.)
    • Dukes B: poor prognosis with age >60, PNI, LVI, tumor <10cm from anal verge, CEA >3.2 ng/ml
    • Dukes C: poor prognosis with male gender, PNI, LVI, tumor <10cm from anal verge, CEA >3.2 ng/ml
    • No survival benefit of RT
  • 1986 PMID 3518912 -- "Postoperative radiotherapy in Dukes' B and C carcinoma of the rectum and rectosigmoid. A randomized multicenter study." (Balslev I, Cancer. 1986 Jul 1;58(1):22-8.)
    • Dukes B: no benefit from RT
    • Dukes C: improved local control at 2 years; time to LR delayed by 1 year
    • Plasma CEA levels influenced by RT

• NSABP R-01 (1977-86) - PMID 3276900 — "Postoperative adjuvant chemotherapy or radiation therapy for rectal cancer: results from NSABP protocol R-01." Fisher B et al. J Natl Cancer Inst. 1988 Mar 2;80(1):21-9.
  • Randomized 3 arms. 555 pts. Dukes' B and C after curative resection. Treated with 1) observation, 2) MOF (5-FU, semustine, vincristine) only, or 3) RT 46-47 Gy
  • Improvement in OS and DFS for chemotherapy only group, but benefit restricted to males.
  • RT resulted in decreased LR from 25% to 16% (p=0.06), but no change in DFS or OS.
  • Surg vs surg + chemo showed no difference in LRR
  • Conclusion: No role for post-op RT; chemo benefit for males.
  • Critique: large non-compliance in RT arm

==Adjuvant Chemo +/- RT ___________________________________________

• NSABP R-02 (1987-92) - PMID 10699069 Full text — "Randomized trial of postoperative adjuvant chemotherapy with or without radiotherapy for carcinoma of the rectum: National Surgical Adjuvant Breast and Bowel Project Protocol R-02." Wolmark N et al. J Natl Cancer Inst. 2000 Mar 1;92(5):388-96.
  • 694 pts. Dukes' B and C after curative resection. Randomized to adjuvant chemotherapy alone or chemotherapy plus RT. All females received 5-FU/LV, whereas males were randomized to MOF vs 5-FU/LV. Median f/u 93 months.
  • Post-op RT resulted in lower LR (13% vs 8%) but did not improve DFS or OS. For male pts, 5-FU/LV resulted in better DFS than MOF but not better OS.
  • Conclusion: No role for post-op RT, although benefit with local control.

• GITSG GI-7175 (1975-80)
  • See below for details
  • Randomized 4 arms. 227 patients. Dukes B and C. (T3-4N0 or N+). Treated with 1) Surgery alone, 2) post-op Chemo (5-FU/MCCNU), 3) post-op RT 40-48 Gy, 4) post-op Chemo-RT 40-44 Gy + 5-FU
  • Significant benefit between chemo-RT arm and observation; please see details below

Adjuvant RT +/- Chemo

• GITSG GI-7175 demonstrated the superiority of post-op chemo-RT over chemo alone, RT alone, or observation.
• NCCTG 79-47-51 demonstrated superiority of post-op Chemo-RT over RT alone.
• ECOG 4276 (negative trial) was only published as abstract.
• Some form of adjuvant chemo-RT is considered standard of care (in patients who already underwent surgical resection, and were found T3-4 or N+ on pathology)

• ECOG EST 4276 (1986-?) - Abstract Only -- "A comparison of post-operative chemotherapy, radiotherapy or combination therapy in potentially curable rectal carcinoma: an ECOG study EST 4276." (Mansour E, Proc ASCO 1991; 10: 154 (A484).)
  • Randomized 3 arms. 237/248 patients. Treated post-op with RT 45/25 vs. 2) chemo (MF) vs. 3) RT + sequential chemo
  • No difference in either local control or OS

• NCCTG 79-47-51 (1980-1986) -- postop RT vs postop chemo-RT
  • Randomized. 204 patients, rectal CA T3-4 or N+, within 12 cm of anal verge. Arm 1) post-op RT vs. Arm 2) post-op chemo-RT. RT given 45/25 + 5.4/3 Gy boost to tumor bed and adjacent LN. Chemo bolus 5-FU bolus + semustine x1 month, then bolus 5-FU 500 mg/m2 concurrent with RT, then 2 months consolidative 5-FU/semustine. Major deviations 9%
  • 7-years; 1991 PMID 1997835 — "Effective surgical adjuvant therapy for high-risk rectal carcinoma." (Krook JE et al. N Engl J Med. 1991 Mar 14;324(11):709-15.). Median F/U 7 years
    • Outcome: 5-year recurrence RT 63% vs. chemo-RT 41% (decreased by 34%, SS). LR 25% vs. 13% (decreased by 47%, SS), DM 46% vs. 29% (SS). 5-year OS ~40% vs. ~55% (decreased by 29%, SS). Reduction in death rate highly significant for LAR (52%), not significant for APR (10%)
    • Toxicity: SBO 5%, median time-to-complication 10 months; overall severe late toxicity 7% (comparable between 45 and 50.4 Gy)
    • Conclusion: Adjuvant chemo-RT superior to RT alone; confirms prior GITSG 7175 results
  • Comment: CRT generally well tolerated, unlike GITSG GI-7175. Adjuvant chemo 1 month first, consolidaation 2 months vs 1.5 years in GITSG. On the basis of this study and GITSG 7175, NIH consensus conference recommended chemo-RT as standard of care for T3-T4 or N+ patients

• GITSG GI-7175 (1975-1980) -- surgery alone vs postop chemo vs postop RT vs postop chemo-RT
  • Randomized, 4 arms. Terminated early due to significant benefit of chemo-RT arm. 227/520 patients. Dukes B and C (T3-4 or N+), distal edge within 12 cm from anal verge. Only R0 resection allowed. Treated with 1) Surgery alone, 2) post-op Chemo (bolus IV 5-FU/M-CCNU), 3) post-op RT 40 or 48 Gy standard fraction, 4) post-op Chemo-RT 40 or 44 Gy standard fraction + 5-FU 500 mg/m2 followed by adjuvant 5-FU/M-CCNU as in chemo alone arm. RT given AP/PA, superior border L4/L5, inferior border included perineum; major deviations in 39%. Consolidative chemo given x1.5 years or until disease progression
  • 7-years, 1985 PMID 2859523, 1985 — "Prolongation of the disease-free interval in surgically treated rectal carcinoma. Gastrointestinal Tumor Study Group." [No authors]. N Engl J Med. 1985 Jun 6;312(23):1465-72. Median F/U 6.7 years.
    • Outcome: Recurrence surgery 55% vs. chemo 46% vs. RT 48% vs. chemo-RT 33% (SS). Benefit of chemo-RT (p=0.009) due to both RT (, better LRC p=0.06) and chemo (better DM p=0.06) components. Initial LRR overall 21%; by arm 24% vs. 27% vs. 20% vs. 11%; initial DM overall 25%. LR in perineum 21%, vagina/uterus 17%, anastomosis 12%, sacrum/coccyx 12%, bladder/prostate 12%. Actuarial OS 36% vs. 46% vs. 46% vs. 56% (p=0.07)
    • Toxicity: Severe nonhematological chemo 15% vs. RT 16% vs. chemo-RT 35%
    • Conclusion: Postoperative chemo-RT significantly better for disease-free survival, with trend to overall survival benefit
  • 8-years, 1986 PMID 3773947, 1986 (No abstract) — "Survival after postoperative combination treatment of rectal cancer." Douglass HO et al. N Engl J Med. 1986 Nov 13;315(20):1294-5. Median F/U 7.8 years
    • Outcome: No new recurrences. Now overall survival benefit for chemo-RT over surgery alone (SS)
    • Conclusion: Postoperative chemo-RT improves overall survival over surgery alone in T3-4 or N+ patients
  • GITSG overview, 1988 PMID 3064191, 1988 — "Adjuvant postoperative radiotherapy and chemotherapy in rectal carcinoma: a review of the Gastrointestinal Tumor Study Group experience." Thomas PR et al. Radiother Oncol. 1988 Dec;13(4):245-52.
    • Severe acute toxic side-effects, but late effects rare in arm 4. Three late deaths (2 enteritis in chemo-RT group, one acute leukemia in chemo group).
    • Conclusions: Compared to surgery alone, postop Chemo-RT improved OS (45% v 27% @ 10yrs), prolonged Time-To-Recurrence, decreased Recurrence Rate (33% v 55%), and decreased LF Rate (10% v 25%).
  • Critique: severely underpowered to show benefit, unequal randomization and not analyzed by intent to treat, semustine x1.5 years increases risk for leukemia

Adjuvant chemo-RT schedules

• Semustine (Methyl-CCNU) was a component of both trials demonstrating benefit, but is known to increase risk of leukemia (4% cummulative risk at 6 years). GITSG 7180 and NCCTG 86-47-51 demonstrated that semustine is not a necessary component of chemo-RT
• In the setting of metastatic CRC, continuous infusion 5-FU is superior to bolus 5-FU (7 PIII trials). Synergistic effects of 5-FU and RT are greatest when there is a continuous exposure to 5-FU for 24-48 hours after RT (PMID 6818194). INT 864751 demonstrated improved OS with infusional 5-FU (70% vs. 60% at 4 years). The improvement was due to increased control of distant mets, with no impact on local control.
• INT 0144 trial evaluated infusion 5-FU vs. bolus 5-FU sandwich around infusion 5-FU/RT. Results at 5 years suggest comparable outcomes

• Intergroup INT-0144 (1994-2000) - bolus vs modulated bolus vs CI 5-FU
  • Randomized. 1917 pts. T3-4 or N+. Follow-on to INT 864751. Randomized to: 1) bolus 5-FU then CI 5-FU/RT then bolus 5-FU vs. 2) CI 5-FU then CI 5-FU/RT then CI 5-FU, vs. 3) bolus 5-FU + LV + levimasole then CI 5-FU/RT then bolus 5-FU + LV + levimasole. RT given 45 Gy to field including presacral and internal illiac LN + 5.4 Gy boost with 2cm margin + optional 3.6 Gy boost to tumor bed if no small bowel in-field
  • 2006 PMID 16877719 -- "Phase III trial of fluorouracil-based chemotherapy regimens plus radiotherapy in postoperative adjuvant rectal cancer: GI INT 0144." (Smalley SR, J Clin Oncol. 2006 Aug 1;24(22):3542-7.). Median F/U 5.7 years
    • 5-year survival: no DFS difference (57-62%), no OS difference (68-71%).
    • LR failure (tumor bed, anastomosis, regional LNs) at first relapse 8% vs. 5% vs. 7%. LR failure in non-T4 patients 5% vs. 3% vs. 5%, and primary surgical treatment without neoadjuvant can be appropriate
    • Toxicity: GI Grade 3/4 41-44%; hematologic Grade 3/4 bolus arm 49-55% vs. CI arm 4%
    • Conclusion: similar survival in all arms, different toxicity profiles and central catheter requirements. LAR reasonable initial resection, since local failure rates only 5% at first failure
  • 2002 ASCO Abstract — "Intergroup 0144 - phase III trial of 5-FU based chemotherapy regimens plus radiotherapy (XRT) in postoperative adjuvant rectal cancer. Bolus 5-FU vs prolonged venous infusion (PVI) before and after XRT + PVI vs bolus 5-FU + leucovorin (LV) + levamisole (LEV) before and after XRT + bolus 5-FU + LV." Smalley SR et al. Proc Am Soc Clin Oncol 22:251, 2003 (abstr 1006)
    • Median f/u 4.6 yrs. No difference in RFS or OS. 5-year estimated OS 72% CI vs. 67% bolus
    • Similar toxicity profiles; less GI toxicity in bolus group, less hematologic toxicity in CI group

• Intergroup INT-0114, 1997 (1990-92) - different modulations of 5-FU
  • 1695 pts. T3-T4 or N+. 2 cycles of infusional 5-FU -> RT + chemo -> 2 cycles of chemo. Randomized to: 1) 5-FU alone, 2) 5-FU + LV, 3) 5-FU + levamisole (levamisole not given during RT), 4) 5-FU + LV + levamisole (levamisole not given during RT). RT dose was 45 Gy + 5.4-9 Gy boost, total 50.4 - 54 Gy.
  • 1997 PMID 9164215 — "Adjuvant postoperative fluorouracil-modulated chemotherapy combined with pelvic radiation therapy for rectal cancer: initial results of intergroup 0114." Tepper JE et al. J Clin Oncol. 1997 May;15(5):2030-9.
  • 2002 PMID 11919230 — "Adjuvant therapy in rectal cancer: analysis of stage, sex, and local control--final report of intergroup 0114." Tepper JE et al. J Clin Oncol. 2002 Apr 1;20(7):1744-50. Median f/u 7.4 yrs.
    • No difference in OS or DFS between groups. 5-yr OS 76% for low risk (T1-2N+ or T3N0), 55% for high risk (T3N+ or Any T4).
  • Other publications: PMID 14701779, 2004 (Impact of hospital procedure volume); PMID 14966087, 2004 (Impact of body mass index); PMID 14512393, 2003 (Analysis of surgical salvage)

• INT/NCCTG 86-47-51, 1994 (1986-90) - PMID 8041415 Full text — "Improving adjuvant therapy for rectal cancer by combining protracted-infusion fluorouracil with radiation therapy after curative surgery." O'Connell MJ et al. N Engl J Med. 1994 Aug 25;331(8):502-7.
  • 660 pts. Stage II-III. 2x2 randomization: type of systemic chemotherapy (5-FU + Semustine or 5-FU) and type of 5-FU infusion with RT (bolus or continuous infusion). Treated with 9 weeks of systemic chemotherapy -> RT with 5-FU -> 2nd cycle of chemotherapy. Higher dose of 5-FU used when given alone than with semustine. RT dose 45 Gy to the pelvis + boost 5.4 - 9 Gy, total 50.4 - 54 Gy. Bolus 5-FU was 500 mg/m2 days 1-3, weeks 1+5 of RT. C.I. 5-FU was 225 mg/m2 daily during RT.
  • Median f/u 46 mo. For C.I., less tumor relapse (37% vs 47%), distant mets (31% vs 40%), time to relapse, and overall survival compared to bolus 5-FU. Decreased tumor relapse by 27%, death by 31%. 4-yr relapse-free survival 63% vs 53%, OS 70% vs 60%. No difference in LR. Increased rate of severe diarrhea for C.I.; higher rate of leukopenia for bolus. No benefit seen for semustine.
  • Conclusion: C.I. 5-FU is superior to bolus 5-FU. Improves DM (extra-pelvic disease) but not LC. Much higher doses of 5-FU were given by continuous infusion than by bolus. No benefit to semustine in addition to 5-FU.

Semustine (M-CCNU)

• NCCTG 86-47-51, 1994 (1986-90) - PMID 8041415
  • Please see above in continuous infusion section.
  • No benefit to semustine in addition to 5-FU.

• GITSG 7180, 1992 (1981-85) - PMID 1548520 — "Radiation therapy and fluorouracil with or without semustine for the treatment of patients with surgical adjuvant adenocarcinoma of the rectum. Gastrointestinal Tumor Study Group." [No authors]. J Clin Oncol. 1992 Apr;10(4):549-57.
  • 210 pts. Pts treated with adjuvant RT/5-FU followed by 12 months of 5-FU +/- Methyl-CCNU (Semustine).
  • Conclusion: Semustine is not an essential component of combined modality therapy.