DeCIDE: A phase III randomized trial of docetaxel, cisplatin, 5-fluorouracil, (TPF) induction chemotherapy in patients with N2/N3 locally advanced squamous cell carcinoma of the head and neck
Reporter: J Taylor Whaley
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 3, 2012
Presenter: Ezra E.W. Cohen, MD Presenter's Affiliation: The University of Chicago, Chicago, IL
Locoregionally advanced head and neck squamous cell carcinoma (H&N SCC) presents clinicians with a challenging scenario with historic poor overall survival of less than 50% at 3 years and local and distant recurrence occurring in a majority of patients.
Several randomized trials initiated in the 1990's demonstrated improvements in local regional control and overall survival with the use of concurrent chemoradiation with platinum-based chemotherapy.
Additionally, a previously published meta-analysis (MACH-NS) demonstrated that concurrent chemoradiation was associated with a non-significant 2.4% improvement in survival over induction chemotherapy followed by radiation alone. This improved survival was due to improved local control; however, induction chemotherapy was associated with decreased metastatic disease.
In the past few years, much interest has been generated regarding the use of induction chemotherapy, followed by concurrent chemoradiation, to potentially improved distant-failure free survival (DFFS); however, to date, there is no level 1A evidence for the use of induction chemotherapy to be followed by chemoradiation.
At ASCO 2009, Hitt et al presented a phase III trial performed in Spain evaluating the use of induction chemotherapy (IC) with docetaxel/cisplatin/5-FU (TPF). The preliminary results presented demonstrated that induction chemotherapy followed by chemoradiation significantly increased time to progression and locoregional control compared with chemoradiation alone in pts with unresectable locally advanced HNSCC.
The University of Chicago has previously published their experience with the use of concurrent chemoradiation utilizing a regimen that consists of Docetaxol, continuous infusion 5-FU, hydroxyurea immediately preceding radiation, and hyperfractionated twice daily radiation at 150 cGy/ fraction (DFHX).
Because induction chemotherapy has been associated with lower distant failure rates in HNSCC, the authors undertook the current study to assess for potential improvements in overall survival. The goal of this trial was to determine whether induction chemotherapy prior to concurrent chemoradiation would improve survival compared to chemoradiation alone.
Materials and Methods
In this phase 3, multi-center, international trial, entry criteria were as follows:
Treatment naive and pathologically confirmed HNSCC;
N2/N3 disease without metastases (AJCC 6 staging);
No prior chemotherapy or oncologic surgeries;
KPS > 70%;
Intact organ and bone marrow function
Patients were randomized to one of the following:
Concurrent chemoradiation with Docetaxol (25 mg/m2), continuous infusion 5-FU (600 mg/m2), hydroxyurea (500 mg BID) immediately preceding radiation, and hyperfractionated twice daily radiation at 150 cGy/ fraction (DFHX).
2 cycles of induction chemotherapy with Docetaxol (75 mg/m2), Cisplatin (75 mg/m2), and continuous infusion 5-FU (750 mg/m2) followed by the same CRT (IC arm).
The primary endpoint of the trial was overall survival. Secondary endpoints included distant-failure free survival (DFSS), failure pattern, and recurrence-free survival (RFS).
The target accrual was initially 400 patients. However, in September 2008 after slow accrual, target accrual was modified to 280 subjects, providing 80% power to detect a hazard ratio HR 0.5 for overall survival.
280 subjects were accrued from 2004-09 with minimum follow-up of 24 months.
142 patients were randomized to the induction arm and 138 patients were randomized to the concurrent chemoradiation arm.
Age, gender, and Karnofsky Performance Status were well balanced. Similarly, tumor size, nodal extent involved, AJCC stage, and smoking status were well balanced. 31% of all patients smoked. Mean age for patients was 57 years old, range 38-82 y/o.
Most common primary tumors site was oropharynx (55% of patients). 14% of patients had oral cavity primary. Most patients had N2 nodal disease.
Of 142 patients randomized to IC, 91% received 2 cycles and 87% continued to CRT.
Treatment adherence during chemoradiation was high for Docetaxel and hydroxyurea, but fewer than 75% of the patients received target dose of 5-FU in both arms.
Radiation was delivered without major deviations in 94% and 95% of patients on IC and CRT arms, respectively.
As expected, the most common grade 3-4 toxicities during induction chemotherapy were febrile neutropenia (9%) and mucositis (8%), and during chemoradiation (both arms combined) they were mucositis (45%), dermatitis (19%), and leukopenia (17%).
Only grade 3-4 leukopenia and neutropenia rates were significantly higher in induction chemotherapy.
In the induction arm, only 9% of patients had a complete response though 55% of patients had a partial response.
IC arm (%)
CRT arm (%)
Distant-Failure Free Survival
Recurrence Free Survival
Cumulative incidence of distant failure
Cumulative incidence of locoregional failure
The outcomes for each arm can be seen in the table above. Although there appears to be trend towards improved recurrence free survival and distant-failure free survival with induction chemotherapy, these were not statistically significant.
The induction chemotherapy arm did demonstrate an improvement in cumulative incidence of distant failure; however, there was no difference between the two arms for overall survival.
Interestingly, cancer death was higher in chemoradiation arm; however, non-cancer death was higher in the induction chemotherapy arm.
There were no significant differences in outcomes for primary site of tumors although oropharynx cancers trended toward improved overall survival.
Additionally, N2c/N3 disease had a trend toward improvement in overall survival with induction chemotherapy; however, the study was not powered to detect differences in these subsets.
In this multi-institutional international study, high survival rates were observed in both arms for patients with locally advanced H&N SCC. Therefore, concurrent chemoradiation with Docetaxol, 5-FU, hydroxyurea, and hyperfractionated radiation appears to be an effective treatment.
Although there was a statistically significant improvement in cumulative incidence of distant metastases in the induction chemotherapy arm, there was not improvement in overall survival.
Further analysis and follow-up may provide insight into why the significant decrease in distant failure did not translate into improved overall survival.
The authors presented a phase III international study evaluating the use of induction chemotherapy followed by concurrent chemoradiation for patients with locally advanced H&N SCC. This is an area of much interest for today's oncologists.
Although previous studies have demonstrated improvements in distant failure, there has not been an improvement in overall survival when compared to concurrent chemoradiation. Unfortunately, due to poor accrual, the results of this trial demonstrate similar results.
At the time of the trial design, HPV had not been demonstrated to be important in H&N cancers. Because of this, patients were not stratified by this factor. From the Bonner et al study, we now know that HPV status is crucial in this disease.
Because of the emergence of HPV as a prognostic factor after the trial began, the results of this study are difficult to interpret. The authors assumed a 3-year overall survival of 50%, which is consistent with historical controls. In recent years, patients have survived much longer, and in fact, the 3-year overall survival in this trial was 73%. This diminished potential differences in overall survival, limiting the interpretation of this data.
Additionally, the induction chemotherapy arm only utilized 2 cycles of TPF. It is uncertain if additional cycles of induction chemotherapy would benefit these patients. The complete response to induction chemotherapy of 8.8% was surprisingly low. However, it is clear from toxicity data that patients receiving induction chemotherapy had increased hematologic toxicity and a 3rd cycle would have been difficult to deliver.
The study was a well-balanced randomized study with appropriate endpoints. However, the study was initially powered to show differences with 400 patients. Due to poor accrual, a revised sample size was amended to reach 280 patients. This change may have diminished difference in the two arms, preventing statistically significant differences from emerging. Specifically, although there may have been differences with induction between patients with HPV and N2c/ N3 disease, the study was not powered to detect these differences.
Overall, the control arm did very well. This is likely related to the high rates of HPV and oropharynx cancers. However, because of the very good outcomes with the control arm, a very large trial would be necessary to detect small differences.
Finally, due to limited follow up (24 months median), additional time and more follow up is needed to further evaluate the outcomes of the trial. The question surrounding the benefit of induction chemotherapy prior to concurrent chemoradation for the treatment of locally advanced HNSCC remains unanswered in full.
Oct 25, 2014 - Several abstracts involving potential biomarkers of prognosis in cancer treatment were presented at a press briefing Nov. 18 at the American Association for Cancer Research -- National Cancer Institute -- European Organisation for Research and Treatment of Cancer International Conference, "Molecular Targets and Cancer Therapeutics," held from Nov. 15 to 19 in Boston.