Departments of 1 Internal Medicine and 2Otolaryngology, Evanston Northwestern Healthcare, Evanston, IL; Departments of 3 Internal Medicine, 4Preventative Medicine, 5Radiation Oncology and 6Otolaryngology, Northwestern University, Feinberg School of Medicine, Chicago, IL and Robert H. Lurie Comprehensive Cancer Center; Departments of 7 Internal Medicine, 8 Radiation Oncology and 9 Otolaryngology, University of Chicago, Chicago, IL and University of Chicago Cancer Research Center; 10 Section of Head and Neck and Thoracic Oncology, M.D. Anderson Cancer Center, Houston, TX; 11 Department of Medicine, John H. Stroger Hospital of Cook County, Chicago, IL; 12 Department of Otolaryngology, Loyola University Stritch School of Medicine, Maywood, IL, USA
* Correspondence to: Dr B. Brockstein, Northwestern University, Feinberg School of Medicine, Evanston Northwestern Healthcare, Section of Oncology/Hematology, 2650 Ridge Ave, Evanston, IL 60201, USA. Tel: +1-847-570-2515; Fax: +1-847-570-2336; Email: b-brockstein{at}northwestern.edu
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Patients and methods: Three hundred and thirty-seven stage IV patients were treated from 1989 to 1998. We compared locoregional and distant recurrence rates, overall survival and progression-free survival from two different treatment strategies: intensive induction chemotherapy followed by split-course chemoradiotherapy (type 1, n=127), or intensified, split-course, hyperfractionated multiagent chemoradiotherapy alone (type 2, n=210). Univariate and multivariate analyses of 12 chosen covariates were assessed separately for the two study types.
Results: The pattern of failure varied greatly between study types 1 and 2 (5-year locoregional failure of 31% and 17% for study types 1 and 2, respectively, P=0.01; 5-year distant failure rate of 13% and 22% for study types 1 and 2, P=0.03). Combined 5-year overall survival was 47% [95% confidence interval (CI) 41% to 53%) and progression-free survival was 60% (95% CI 55% to 66%). Both treatment strategies yielded similar survival rates. Poor overall survival and distant recurrence were best predicted by advanced nodal stage. Locoregional recurrence was extremely rare for patients with T0T3 tumor stage, regardless of lymph-node stage.
Conclusions: This analysis suggests that pattern of failure in primary head and neck cancer may be dependent upon treatment strategy. Randomized clinical trials of induction chemotherapy are warranted as a means to determine if a decrease in distant metastases can lead to an increase in survival rates in the setting of effective chemoradiotherapy for locoregional control. Additionally, this analysis provides impetus for randomized clinical trials of organ preservation chemoradiotherapy in sites outside the larynx and hypopharynx.
Key words: chemoradiotherapy, head and neck cancer, induction chemotherapy, patterns of failure, prognostic factors
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Historically, locoregional recurrences have been more common after treatment of advanced HNC with CRT than are distant metastases: 50% versus 1520% [7
9
]. Recent studies using more intensive CRT have reported locoregional and distant failure rates that are approximately equal [10
12
]. This may be a direct result of improved locoregional treatment programs with suboptimal influence of the chemotherapy upon distant disease, and this has emphasized the need for attention to treatment of distant micrometastatic disease.
The goals of this analysis were to analyze, by study type, changes in patterns of failure and prognostic factors for locoregional and distant recurrence, progression-free and overall survival using two different treatment strategies in 337 patients treated over a 9-year period. Additionally, by study type, we assessed this overall intensive treatment strategy in terms of the ultimate end points of survival and tumor control in the setting of organ preservation.
Our trials all utilized the FHX (5-fluorouracil, hydroxyurea and radiotherapy) backbone as initially reported by Vokes et al. [13], based on fundamental principles as established by Byfield et al. [14
], Taylor et al. [15
] and others. Our first two trials, from 1989 to 1993 (type 1 trials, Figure 1 and Table 1), utilized induction chemotherapy, followed by concomitant, split-course, FHX chemoradiotherapy [16
, 17
]. By 1993, induction chemotherapy had been shown to not improve survival in randomized trials. Since locoregional failures predominated, induction chemotherapy was deleted from the subsequent three regimens (type 2 trials). These trials, conducted from 1993 to 1998, focused on further improvement in locoregional control and utilized concomitant split-course hyperfractionated RT with a third drug added to FHX [10
, 11
, 18
].
|
|
![]() |
Patients and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Protocols
The details of the five treatment protocols have been published previously [10, 11
, 16
18
] and are detailed in Table 1 and Figure 1.
Protocols 1 and 2 were nearly identical (type 1 studies) [16, 17
]. Three cycles of PFLI induction chemotherapy were given prior to FHX chemoradiation (single fraction RT, 200 cGy daily). FHX cycles were repeated every 14 days to complete a total RT dose of 6000 cGy for resected patients and
7000 cGy for all others. Surgery, when used in these two trials, took place most commonly between induction chemotherapy and CRT and consisted almost completely of limited primary site resection and neck dissection, although a few patients had more definitive surgery.
Protocols 3, 4 and 5 (type 2 studies) used no induction chemotherapy. Instead, a third agent (cisplatin, protocol 3, and paclitaxel, protocols 4 and 5) was added to FHX, and hyperfractionated RT (150 cGy twice daily) was used. Cycles were repeated every 14 days to complete a total RT dose of 6000 cGy for patients without gross disease and 7000 cGy for all others. Only limited primary site resections (such as laser excision/debulking of a tumor) were utilized (although rarely), except as salvage. Neck dissection was recommended for patients with N2 or N3 disease, generally after CRT.
Protocol study group and database
In total, 394 patients were treated on the five protocols. A uniform group of 337 patients with stage IV, M0, squamous cell carcinoma (SCCA) of the oral cavity, oropharynx, hypopharynx, larynx or neck with unknown primary site was analyzed for this study. The reasons for exclusion of 57 patients were: stage III disease (18 stage III base of tongue or pyriform sinus who were eligible for studies); non-squamous histology (22 lymphoepithelioma/nasopharynx site, and two mucoepidermoid carcinoma); other sites (14 mostly paranasal sinus); and one patient with missing data.
All data were recorded and managed at treating sites, and regularly sent to the central institution (University of Chicago) to form the five databases. At the time of analysis for publication of each protocol, data verification was performed by the physician investigators, data managers and statisticians. At the time of this analysis, all outcome data were updated to last patient contact (<6 months in almost all cases) or death. Individual demographic and outcome data were re-verified by physician and data management review. The relevant prognostic factors, baseline characteristics, and toxicity and outcome measures of interest were then reduced to a single dataset.
Statistical methods, definition of survival and prognostic factor analysis
Locoregional progression-free survival (LRPFS), distant recurrence-free survival (DRFS), overall survival (OS) and progression-free survival (PFS) were analyzed using KaplanMeier product limit curves. Duration of all types of survival was counted from the on-study date. In the LRPFS and DRFS analyses, progression was defined as any locoregional or distant progression, respectively (12 patients had simultaneous locoregional and distant recurrence; five patients had local relapse followed by distant relapse, and none had distant followed by locoregional relapse). In the PFS analysis, progression was defined as locoregional progression, distant progression, toxic death or death from disease. For PFS, surviving patients and those dying of unrelated or intercurrent illnesses were censored at last follow-up or death.
Univariate analyses and multivariate analyses were conducted. Protocol types 1 and 2 were analyzed separately. For the univariate analyses, the log-rank test was used to compare KaplanMeier curves. Multivariate analysis was then conducted using Cox proportional hazards regression on those variables with P 0.05 in the univariate analyses. Results are expressed as hazard ratios relative to a chosen reference group, with corresponding P values.
Prognostic variables
Prognostic variables were chosen from the dataset based on clinical relevance, potential for influence on outcome, or previous reports of prognostic relevance. Chosen covariates are defined in Table 2. T stage was grouped as T03 versus T4, since locoregional relapse in T03 subgroups was exceedingly rare. N stage was analyzed by sub-stage and also dichotomized, to improve power to detect influence of N stage, as N0N2B versus N2CN3. Treating institution was reduced from seven to four groups based on institutional affiliations, and very low treatment numbers in three institutions.
|
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
|
|
Multivariate analysis. Multivariate analysis was performed for study type 2 only, since only single univariate parameters were significant for study type 1 outcome measures in univariate analyses, and are shown in Table 5. The only significant predictor of favorable OS in these stage IV patients was early N stage [N0N2B versus N2CN3, P = 0.035, hazards ratio (HR) 1.51]. Locoregional failure was strongly correlated with T4 stage (P = 0.002, HR 6.79). Notably, this was independent of N stage, as patients with T03 and any N stage still only rarely relapsed (three of 91) in the primary site or neck. Interestingly, treating institution remained significant after adjusting for other factors (P = 0.042). Distant failure was predicted only by advanced N stage (P = 0.009, HR 2.26).
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
This analysis demonstrated that there was a clear reversal in the pattern of failure from study type 1 to study type 2. The improvement in locoregional control in study type 2 is likely a result of hyperfractionated RT and/or the addition of a third sensitizing chemotherapy drug. The increase in distant failure in study type 2 is likely due to the deletion of induction chemotherapy. Confounding factors which would alter the rate of distant metastases would have led to more distant metastases in type 1 patients (type 1 group had a higher N stage and were treated earlier in the CT scan era when more small distant metastases would have been missed pre-treatment). Improved locoregional control, however, may have allowed time for recognition of distant metastases. Notably, the three chemotherapy drugs used simultaneously with RT (but at lower doses and fewer cycles than with induction chemotherapy) did not compensate for the deletion of induction chemotherapy in decreasing distant metastases.
FHX-based chemoradiation as given on these protocols resulted in promising survival in this patient population. Any negative effects of split-course-related treatment delay were outweighed by the efficacy and intensity of the treatment. These data suggest that intensive CRT might function as a substitute to surgery for resectable squamous cell HNC patients when appropriately delivered at experienced institutions. Several other recently published protocols have also suggested this finding in resectable patients [12, 19
22
]. Based on these observations, randomized studies should be performed to compare CRT to primary surgical therapy, as well as to establish an optimal CRT regimen.
We performed prognostic factor analyses separately for protocol types 1 and 2. Since the pattern of significant variables was different for the two types of studies, interaction terms between study type and prognostic variable would need to be in any analysis that combined both types of studies. More specific information is extracted from these data if analyses are separated by study type (1 versus 2). Age and nodal status differed significantly between the two types of studies. To determine whether these factors were confounding variables that may explain outcome differences, a multivariate analysis of all patients, including study type, age and nodal status was performed. This indicated that study type remained statistically significant for distant recurrences (P=0.027) but became marginally significant for locoregional recurrences (P=0.068). These reductions in statistical significance were due to age and not nodal status.
The only significant predictor of OS in protocol type 2 was N stage, dichotomized between N0N2B and N2CN3. Neck failure in type 2 studies was very rare, and therefore advanced N stage predicted death from causes other than neck failure, likely distant metastases. In fact, the difference in survival between the N0N2B and N2CN3 groups parallels in timing and magnitude the differences in distant metastasis rate between these two groups (data not shown). This is an important observation, as it suggests that lowering the rate of distant metastases will likely directly improve survival. Other series have suggested advanced N stage as a poor prognostic factor for survival [2325
] in locoregional stage IV patients. However, locoregional failure, the traditional mode of failure in advanced HNC, may have contributed to the deaths of the patients in these other series.
Locoregional recurrence in type 2 studies was predicted by T stage. Only three of 91 patients with T0T3 had locoregional recurrences in the primary site or neck (all N2 by definition of stage IV), in contrast to 27% of patients with T4 tumors. This difference was present but less pronounced in the type 1 patients. This high rate of locoregional control represents an improvement over most series with standard surgery and radiation, despite the theoretical detrimental effect of the requisite scheduled treatment breaks. The true benefit or detriment of CRT versus surgery for these subsets remains to be determined in randomized studies.
Multivariate analysis showed that treating institution also predicted for locoregional failure. Further analysis showed that this poor outcome was accounted for by one institution. Others have reported similar findings from multi-institutional studies [26]. The cause of this difference is unclear, but it suggests that experience, treatment reproducibility and equipment optimization are critical.
Finally, we found that distant metastasis was best predicted for by advanced lymph-node stage. When we specifically assessed failure patterns in T0T3, N2CN3 patients, we found a striking difference in distant progression rates between protocol type 1 and 2 patients (Figure 6). Since the T0T3 patients had near uniform locoregional control, this suggests that the induction chemotherapy given in protocol type 1 accounted for this marked difference in metastases. These results are similar to those of others who have studied risk factors for distant metastases in a variety of stages and treatment settings [23, 26
28
]. More importantly, however, our data comparing type 1 and type 2 studies suggest that induction chemotherapy may reduce the risk of developing distant metastases by 3040%. This is almost identical to the results of six large randomized trials [2
, 29
33
] of induction chemotherapy, and one large phase II study [22
], which have assessed this issue of induction chemotherapy. Initial results from our two most recent trials that re-introduced induction chemotherapy to type 2 CRT also suggest the same reduction in rates of metastatic disease [34
, 35
].
|
We conclude that CRT may be highly effective at improving survival and allowing anatomic organ preservation when given with appropriate intensity by a skilled multidisciplinary group. Ongoing studies by our group and others are assessing functional organ preservation and quality of life as other important end points. Patients with T0T3 tumor stage have near-uniform locoregional control independent of N stage, and even T4 patients have a 73% rate of locoregional control at 5 years. Advanced N stage predicts for increased distant metastases and lower OS. Induction chemotherapy may improve distant control, and strategies such as the reintroduction of induction chemotherapy should be assessed as a means of improving survival in light of the now excellent locoregional control. The optimal induction chemotherapy regimen has not been clearly defined, and less intensive regimens than PFLI used in this study appear to be no less effective [2, 22
, 29
35
]. Finally, it should be remembered that not all CRT is the same, and comparative, randomized studies are needed.
![]() |
Acknowledgements |
---|
Received for publication January 20, 2004. Revision received March 24, 2004. Accepted for publication March 24, 2004.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2. The Department of Veterans Affairs Cancer Study Group. Induction chemotherapy plus radiation compared with surgery plus radiation in patients with advanced laryngeal cancer. N Engl J Med 1991; 324: 16851690.[Abstract]
3. Lefebvre JL, Chevalier D, Luboinski B et al. Larynx preservation in pyriform sinus cancer: preliminary results of a European Organization for Research and Treatment of Cancer phase II trial. J Natl Cancer Inst 1996; 88: 890899.
4. Forastiere AA, Goepfert H, Maor M et al. Concurrent chemotherapy and radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med 2003; 349: 20912098.
5. Forastiere A, Koch W, Trotti A, Sidransky D. Head and neck cancer. N Engl J Med 2001; 345: 18901900.
6. Pignon JP, Bourhis J, Domenge C, Designe L. Chemotherapy added to locoregional treatment for head and neck squamous cell carcinoma: three meta-analyses of updated individual patient data. Lancet 2000; 355: 949955.[CrossRef][ISI][Medline]
7. Koness RJ, Glicksman A, Liu L et al. Recurrence patterns with concurrent platinum based chemotherapy and accelerated hyperfractionated radiotherapy in stage III and IV head and neck cancer. Am J Surg 1997; 174: 532535.[CrossRef][ISI][Medline]
8. Garden AS, Pajek TF, Vokes E et al. Preliminary results of RTOG 9703a phase II randomized trial of concurrent radiation (RT) and chemotherapy for advanced squamous cell carcinoma of the head and neck. Proc Am Soc Clin Oncol 2001; 20: 891 (Abstr).
9. Robbins KT, Kumar P, Regine WF et al. Efficacy of targeted supradose cisplatin and concomitant radiation therapy for advanced head and neck cancer: the Memphis Experience. Int J Radiation Oncol Biol Phys 1997; 38: 263271.[ISI][Medline]
10. Vokes EE, Kies M, Haraf DJ et al. Concomitant chemoradiotherapy as primary therapy for locoregionally advanced head and neck cancer. J Clin Oncol 2000; 18: 16521661.
11. Kies MS, Haraf DJ, Rosen F et al. Concomitant infusional paclitaxel and fluorouracil, oral hydroxyurea, and hyperfractionated radiation for locally advanced squamous head and neck cancer. J Clin Oncol 2001; 19: 19611969.
12. Adelstein DJ, Saxton JP, Lavertu P et al. Maximizing local control and organ preservation in stage IV squamous cell head and neck cancer with hyperfractionated radiation and concurrent chemotherapy. J Clin Oncol 2002; 20: 14051410.
13. Vokes EE, Panje WR, Schilsky R et al. Hydroxyurea, fluorouracil, and concomitant radiotherapy in poor-prognosis head and neck cancer. J Clin Oncol 1989; 7: 761768.[Abstract]
14. Byfield JE, Sharp TR, Frankel SS et al. Phase I and II trial of five-day infused 5-fluorouracil and radiation in advanced cancer of the head and neck. J Clin Oncol 1984; 2: 406413.[Abstract]
15. Taylor SGIV, Murthy AK, Caldarelli DD et al. Combined simultaneous cisplatin/fluorouracil chemotherapy and split course radiation in head and neck cancer. J Clin Oncol 1989; 7: 846856.[Abstract]
16. Vokes EE, Kies MS, Haraf DJ et al. Induction chemotherapy followed by concomitant chemoradiotherapy for advanced head and neck cancer: impact on the natural history of the disease. J Clin Oncol 1995; 13: 876883.[Abstract]
17. Kies MS, Haraf DJ, Athanasiadis I et al. Induction chemotherapy followed by concurrent chemoradiation for advanced head and neck cancer: improved disease control and survival. J Clin Oncol 1998; 16: 27152721.[Abstract]
18. Rosen FR, Haraf DJ, Kies MS et al. Multicenter randomized phase II study of paclitaxel (1-hour infusion), fluorouracil, hydroxyurea and concomitant twice daily radiation with or without erythropoietin for advanced head and neck cancer. Clin Cancer Res 2003; 9: 16891697.
19. Glicksman AS, Wanebo HJ, Slotman G et al. Concurrent platinum based chemotherapy and hyperfractionated radiotherapy with late intensification in advanced head and neck cancer. Int J Radiat Oncol Biol Phys 1997; 39: 721729.[ISI][Medline]
20. Leyvraz S, Pasche P, Bauer J et al. Rapidly alternating chemotherapy and hyperfractionated radiotherapy in the management of locally advanced head and neck carcinoma; four year results of a phase I/II study. J Clin Oncol 1994; 12: 18761885.[Abstract]
21. Adelstein DJ, Lavertu P, Saxton JP et al. Mature results of a phase III randomized trial comparing concurrent chemoradiotherapy with radiation therapy alone in patients with stage III and IV squamous cell carcinoma of the head and neck. Cancer 2000; 88: 876883.[CrossRef][ISI][Medline]
22. Clark JR, Busse PM, Norris CM Jr et al. Induction chemotherapy with cisplatin, fluorouracil, and high-dose leucovorin for squamous cell carcinoma of the head and neck: long-term results. J Clin Oncol 1997; 15: 31003110.[Abstract]
23. Cerezo L, Millan I, Torre A et al. Prognostic factors for survival and tumor control in cervical lymph node metastases from head and neck cancer. A multivariate study of 492 cases. Cancer 1992; 69: 12241234.[ISI][Medline]
24. Mamelle G, Pampurik J, Luboinski B et al. Lymph node prognostic factors in head and neck squamous cell carcinomas. Am J Surg 1994; 68: 494498.
25. Fortin A, Couture C, Doucet R et al. Does histologic grade have a role in the management of head and neck cancers? J Clin Oncol 2001; 19: 41074116.
26. Taylor SG, Murthy AK, Vannetzel P et al. Randomized comparison of neoadjuvant cisplatin and fluorouracil infusion followed by radiation versus concomitant treatment in advanced head and neck cancer. J Clin Oncol 1994; 12: 385395.[Abstract]
27. Leemans CR, Tiwari R, Nauta JJ et al. Regional lymph node involvement and its significance in the development of distant metastases in head and neck carcinoma. Cancer 1993; 71: 452456.[ISI][Medline]
28. Vikram B, Strong EW, Shah JP, Spiro R. Failure at distant sites following multimodality treatment for advanced head and neck cancer. Head Neck Surg 1984; 6: 730733.[ISI][Medline]
29. Laramore GB, Scott CB, Al-Sarraf M et al. Adjuvant chemotherapy for resectable squamous cell carcinomas of the head and neck: report on Intergroup study 0034. Int J Radiat Oncol Biol Phys 1992; 23: 705713.[ISI][Medline]
30. Paccagnella A, Orlando A, Marchiori C et al. Phase III trial of initial chemotherapy in stage III or IV head and neck cancers: a study by the Gruppo di Studio Sui Tumori della Testa e del Collo. J Natl Cancer Inst 1994; 86: 265272.[Abstract]
31. Head and Neck Contracts Program. Adjuvant chemotherapy for advanced head and neck squamous carcinoma. Final report of the head and neck contracts program. Cancer 1987; 60: 301311.[ISI][Medline]
32. Schuller DE, Metch B, Stein DW et al. Preoperative chemotherapy in advanced resectable head and neck cancer: final report of the Southwest Oncology Group. Laryngoscope 1988; 98: 12051211.[ISI][Medline]
33. Domenge C, Hill C, Lefebvre JL et al. Randomized trial of neoadjuvant chemotherapy in oropharyngeal carcinoma. Br J Cancer 2000; 83: 15941598.[CrossRef][ISI][Medline]
34. Vokes EE, Stenson K, Rosen F et al. Weekly carboplatin and paclitaxel followed by concomitant paclitaxel, fluorouracil, and hydroxyurea chemoradiotherapy: curative and organ-preserving therapy for advanced head and neck cancer. J Clin Oncol 2003; 21: 320326.
35. Haraf DJ, Rosen FR, Stenson K et al. Induction chemotherapy followed by concomitant TFHX chemoradiotherapy with reduced dose radiation in advanced head and neck cancer. Clin Cancer Res 2003; 9: 59365943.