Postoperative irradiation with or without concomitant chemotherapy for locally advanced head and neck cancer
Reviewed by: Stephen Z. Sack, MD PhD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 27, 2004
Authors: Bernier J, Domenge C, Ozsahin M, Matuszewska K, Lefebvre JL, Greiner RH, Giralt J, Maingon P, Rolland F, Bolla M, Cognetti F, Bourhis J, Kirkpatrick A, van Glabbeke M Source:N Engl J Med. 2004 May 6;350(19):1945-52.
In 1994 the European Organization for Research and Treatment of Cancer (EORTC) began trial 22931 to test the hypothesis that adjuvant chemoradiation therapy improved progression free survival, overall survival and local-regional control more than radiation therapy alone in Stage III or IV Head and Neck squamous cell carcinomas.
These results were previously described in an interim report (Ann Oncol 2002), which, with a medium follow up of only 34 months already suggested improvements in outcome with combined therapy.
This paper was published back-to-back with the American study from the Radiation Therapy Oncology Group (RTOG) trial 95-01 and together these papers strongly support a change in the accepted standard of care from radiation therapy alone to combined chemoradiation therapy for advanced head and neck squamous cell carcinomas.
After surgery, patients were centrally randomized with center and tumor site being used as stratification factors.
Primary endpoint was progression free survival, which is defined as time from randomization to progression or death.
Secondary endpoints included overall survival, local-regional control, metastasis, second primary tumor, and late toxicity.
Age > 18 years old and less than 70 years old with good performance status (0,1, or 2), adequate pretreatment lab values and a hemoglobin ≥ 11 (Note: Low hemoglobin levels are associated with worse outcomes, and there has been much recent interest in the possible adverse impact to survival of exogenous colony stimulating factors e.g. Erythropoietin)
Site: Oral cavity, oropharynx, hypopharynx or larynx (Note: T3N0 larynx excluded due to goal of preservation of function and thus treated with chemoradiation therapy)
Stage pT3-pT4 or Stage 1-2 with N2-3 or low risk patients with adverse prognostic indicators (e.g. extranodal spread, positive margins, perineural invasion, or vascular tumor emboli)
Staging CT was optional
Surgery was considered close if within 5mm of surgical margin.
Radiation given postoperatively in using conventional fractionation of 2 Gy/day (Monday through Friday) using linear accelerators of 4-6 MV with the use of isocentric techniques. The primary site and all draining lymph nodes received 54 Gy in 27 fractions with a 12 Gy boost at sites of high risk or inadequate margins.
Chemotherapy consisted on Cisplatinin 100 mg per square meter of body surface area days 1, 22 and 43.
Median follow-up 60 months (5 years).
As compared to the concurrently published RTOG study, this European study has a higher proportion of patients with adverse prognostic indicators such as margin status (28% in EORTC vs 17-19% in RTOG 95-01), worse tumor differentiation (19% in EORTC vs 7% in RTOG 95-01), and more hypopharynx cancers(20% in EORTC vs 7-12% in RTOG 95-01). Also a larger proportion of these patients (32%) had greater than a 6 week delay prior to starting postoperative therapy (Note: Data suggests that outcomes worsen for package times less than 11 wks, vs 11-13 wks, vs greater than 13 wks, Kian Ang et al., IJROB, 2001). Prolongation of treatment/treatment interruptions were noted in 81 of the patients who received e 60 Gy.
Chemotherapy compliance was yet another issue, with 10% of patients on that arm of the study refusing chemotherapy with a total of 36% of patients on that arm not receiving all 3 cycles of chemotherapy, and only 49% of patients receiving the planned three courses without any delay or dose reduction.
With a median follow-up of 5 yrs, there was a statistically significant improvement in progression-free survival, overall survival, and local regional control for the combined therapy arm vs the radiotherapy alone arm.
Estimated median duration of progression-free survival was 23 months in the radiotherapy group and 55 months in the combined-therapy group.
The estimated median time to death was 32 months in the radiotherapy group and 72 months in the combined-therapy group.
The estimated 5-year cumulative incidence of local or regional relapses was 31% in the radiotherapy group and 18% in the combined-therapy group.
The incidence of metastases and of second primary tumors was not statistically significant in the two groups.
There was significantly worse Grade 3 or higher acute toxicity in the combined-therapy group with respect to mucosal toxicity as well as chemotherapy related toxicities such as severe leucopenia, severe granulocytopenia, as well as nausea and vomiting. The long term toxicity profile between the two groups was not statistically different.
Concurrent chemotherapy and radiotherapy significantly increased progression-free and overall survival with 5 yr estimates of 47% and 53% respectively in the combined-therapy group vs 36% and 40% in the radiotherapy alone group.
Combined therapy did not reduce the probability of distant relapse.
Laryngeal cancer trials have foreshadowed these two papers with respect to the role of combined chemoradiation treatment. Prior to 1990, laryngeal cancer was commonly managed with total laryngectomy, however, the Veterans Affairs (VA) trial (NEJM, 1991) compared total laryngectomy with induction chemotherapy and radiation therapy. The chemotherapy regiment consisted of Cisplatin and fluorouracil, and responders received definitive radiation therapy after three cycles of chemotherapy. Two thirds of those treated with combined chemoradiation benefited from organ preservation and showed equivalent survival to those patients treated with total laryngectomy. This trial provided the basis for routine use of neoadjuvant chemotherapy in hopes of organ preservation. In 1991, the Radiation Therapy Oncology Group (RTOG) initiated a randomized study that built on this foundation of neoadjuvant chemotherapy for laryngeal cancer. Results show that the highest rate of organ preservation was in patients treated with concomitant RT and chemotherapy with no significant difference in overall survival. With this understanding of the background for the presumption that combined chemotherapy and radiation may be more effective in head and neck cancers, in 1991, a European and an American trial were undertaken to evaluate if concomitant Cisplatin and radiation were beneficial in locally advanced head and neck cancers.
Prior to the publication of the results of these studies, another study (Intergroup study 0099 [Al-Sarraf et al., JCO 1998]) showed an improvement in combined therapy for advanced nasopharyngeal carcinomas. Intergroup study 0099 results argued for a change in the standard of care, however due to poor results in the radiotherapy alone arm and the small size of the trial (150 patients analyzed), many questioned the strength of these data.
The preliminary report from the American study (RTOG 95-01) showed an improvement in local regional control but there was no statistically significant difference in 2 year disease-free survival or overall survival. An interim report of the this paper (EORTC 22931, Ann Oncol 2002) showed statistically significant improvements in local regional control, disease-free survival and overall survival. Thus the uncertainty created by the divergent results from the interim analyses of RTOG 95-01 and EORTC 22931 left doubt as the role of combined chemoradiation.
An update to the American study (RTOG 95-01, NEJM, 2004) was published back to back with this European study (EORTC 22931) and both suggest a paradigm shift is in order. The issues raised by the interim analysis results seem to be resolved after longer follow-up. Now, both groups find a statistically significant survival improvement with the addition of chemotherapy to definitive radiation.
Many argued that the EORTC results initially showed statistically significant improvement with combined therapy due to the different nature of the patients enrolled in that study as compared to those of the RTOG 95-01 study. In generally, the EORTC study patients has worse prognostic indicators such as margin status, degree of differentiation, and percent of patients with poorer prognosis (e.g. hypopharynx cancer). Those in opposition to changing patterns of care argued that chemotherapy was thus only significant for high risk patients with locally advanced disease. Now, however, results seem to suggest the addition of chemotherapy may benefit selected patients with head and neck locally advanced squamous cell carcinomas.
In this study, only 49% of patients receiving the planned three courses of chemotherapy without any delay or dose reduction, and thus the benefit of combined chemoradiation may be understated in this study.
Caution must however be exercised before undertaking such a regimen. The extent of benefit from combined therapy at particular head and neck sites cannot be reliably assessed in these studies. Furthermore, many find combined chemoradiation with cisplatin is difficult to tolerate, often requiring delays or interruptions in radiation therapy. The unconfirmed assumption is often made that treating with carboplatin may show similar benefits while being better tolerated by patients. With either chemo-therapeutic regimen, the increased toxicity of a combined chemoradiation therapy is more difficult for the patient to tolerate. With head and neck cancers, delays in treatment are thought to have adverse affects on outcome, and thus risks and benefits must be weighed. It thus becomes an important clinical decision to determine if the patient can tolerate a combined chemoradiation regiment, and whether or not the toxicity profile is acceptable.