Larynx Preservation and Tumor Control in Stage III and IV Laryngeal Cancer: A Three-Arm Randomized Intergroup Trial; RTOG 91?11

Reviewer: Roberto Santiago, MD
Last Modified: October 8, 2002

Presenter: Moshe Maor
Presenter's Affiliation: M. D. Anderson Cancer Center, Houston, TX
Type of Session: Plenary


  • Previously, the Veterans Administration evaluated whether irradiation of patients who respond to induction chemotherapy could safely preserve the larynx in patients with stage III and low-volume stage IV glottic and supraglottic cancers.
  • The study showed that ~85% of patients had a response to induction cisplatin (CDDP) and 5FU chemotherapy.
  • When those patients underwent radiotherapy (RT) the larynx was preserved in 64% at 2 years.
  • In addition, this approach significantly decreased the incidence of distant metastases when compared to laryngectomy.
  • Although more local-regional recurrences were observed in the chemoradiotherapy arm, the overall survival was not significantly different than that obtained with surgery.
  • A high proportion of the local-regional recurrences were effectively salvaged.
  • The present study evaluates whether there is a more effective way of delivering the chemoradiotherapy and whether the chemotherapy (CT) is essential for effective laryngeal preservation in the treatment of stage III and low-volume stage IV glottic and supraglottic cancers.
  • The 3 arms of the study are induction CT followed by RT (control) versus concomitant CT and RT versus RT alone.
  • Endpoints for evaluating the trial were laryngectomy free-survival (LFS), larynx preservation rate (LPR), patterns of relapse, treatment-related adverse events, and overall survival.

Materials and Methods

  • Between August 1992 and May 2000, 547 patients were enrolled.
  • Patients were eligible for the study if they had a new diagnosis of a potentially resectable stage III or IV squamous carcinoma of the glottic or supraglottic region; patients with a T1 or a high-volume T4 tumor were excluded from the study.
  • Sixty eight percent of the 517 patients had supraglottic cancer.
  • All patients had N2 disease or higher.
  • All had good performance status as well as adequate blood counts and renal function.
  • Patients were randomly assigned to 3 treatment arms.
  • The arms were well balanced regarding patient and tumor characteristics.
  • Patients in arm 1 (n=173) received 3 cycles of induction CDDP 100 mg/m2 once and 5FU 1000 mg/m2 /day for 5 days every 3 weeks. In responding patients, this treatment was followed by 70 Gy of RT in 35 fractions for 49 days.
  • Patients in arm 2 (n=172) received concurrent CDDP 100 mg/m2 on days 1, 22, and 43 of RT (70 Gy/35 fractions for 49 days).
  • Patients in arm 3 (n=172) received RT only at 70 Gy/35 fractions for 49 days.
  • Patients with a neck node 3 cm or with multiple neck nodes underwent a neck dissection 8 weeks after completion of therapy.


  • The LPR was significantly greater in arm 2 compared with arms 1 and 3 and there was no significant difference between arm 1 and arm 3. The number of laryngectomies at 2 years was 43, 21, and 49 for the induction, concomitant and RT arms respectively.
  • Loco-regional control at 2 years for patients in arm 2 (78%) was significantly better than either arm 1 (61%), or arm 3 (56%), p < 0.01. Loco-regional control at 2 years was not significantly different in arm 1 compared to arm 3.
  • At 2 years, there was no significant difference in LFS between the arm 1 and either of the experimental arms. However, LFS on arm 2 was significantly better than on arm 3 (p = 0.018).
  • Acute grade 4 and 5 toxicity was 31%, 21%, and 5% in treatment arms 1, 2, and 3 respectively (<0.0001). Acute toxicity was not significantly different in arm 1 compared to arm 2.
  • Late grade 4 and 5 toxicity was 9%, 8% and 10% in the 3 arms (not significant).
  • Ten patients in the study died of treatment toxicity; 5 died in the induction CT arm and 5 died in the concomitant CT and RT arm.
  • There was no significant difference in the rates of distant metastases in arm 1 compared to arm 2 or arm 3. However, arm 2 had significantly fewer distant metastases than arm 3.
  • Overall survival was very similar for the patients in the 3 treatment arms (75% at 2 years).
  • All three arms exhibited high rates of failure salvage.
  • Most events were observed within the first 2 years of follow up.

Author's Conclusions

  • Concomitant CT and RT was superior for larynx preservation and loco-regional control compared with induction CT followed by RT and RT alone.
  • Patients on concomitant treatment had a superior LFS compared to patients treated with RT alone and there was no significant difference between the induction CT arm and the RT alone arm using these criteria.
  • Better local control with the concomitant treatment arm did not improve survival.

Clinical/Scientific Implications

  • In appropriately selected patients, concurrent CDDP-based chemoradiotherapy is, at the moment, the most effective approach for laryngeal preservation albeit with higher acute toxicity.
  • Questions that remain unanswered include: 1. Can we identify favorable subsets (i.e. patients with small volume disease, < 3.5 cm3 by CT) that could be treated by RT alone? 2. Can we identify unfavorable subsets doomed to fail that could benefit from upfront laryngectomy? 3. Can these results be improved on with novel RT fractionation, chemotherapy and imaging schemes? 4. Can we modify the tumor environment (i.e. anemia, biologically targeted agents) to our advantage?
  • Last but not least, extreme caution must be taken not to extrapolate these results to other head and neck sites with poorer chances of failure salvage.

Oncolink's ASTRO Coverage made possible by an unrestricted Educational Grant from Ortho Biotech.