A Phase I/II trial of concurrent docetaxel and radiation after induction chemotherapy in patients with poor prognosis squamous cell carcinoma of the head and neck


  • Roy B. Tishler M.D., Ph.D.,

    Corresponding author
    1. Department of Radiation Oncology, Beth Israel Deaconess Medical Center, Boston, Massachusetts
    • Department of Radiation Oncology, Beth Israel Deaconess Medical Center, 330 Brookline Avenue, Boston, MA 02215
    Search for more papers by this author
    • Fax: 617-667-4990

  • Charles M. Norris Jr. M.D.,

    1. Head and Neck Oncology Program, Dana Farber Cancer Institute, Boston, Massachusetts
    2. Department of Surgery, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts
    Search for more papers by this author
  • A. Dimitrios Colevas M.D.,

    1. Head and Neck Oncology Program, Dana Farber Cancer Institute, Boston, Massachusetts
    2. Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts
    Current affiliation:
    1. Investigational Drug Branch, National Cancer Institute/Cancer Therapy Evaluation Program (CTEP), Rockville, MD
    Search for more papers by this author
  • Carolyn C. Lamb M.D.,

    1. Department of Radiation Oncology, Beth Israel Deaconess Medical Center, Boston, Massachusetts
    Search for more papers by this author
  • Daniel Karp M.D.,

    1. Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts
    Search for more papers by this author
  • Paul M. Busse M.D., Ph.D.,

    1. Department of Radiation Oncology, Beth Israel Deaconess Medical Center, Boston, Massachusetts
    Search for more papers by this author
  • Asa Nixon M.D.,

    1. Department of Radiation Oncology, Beth Israel Deaconess Medical Center, Boston, Massachusetts
    Search for more papers by this author
  • Robert Frankenthaler M.D.,

    1. Department of Surgery, Beth Israel Deaconess Medical Center, Boston, Massachusetts
    Search for more papers by this author
  • Bernadette Lake-Willcutt B.A.,

    1. Head and Neck Oncology Program, Dana Farber Cancer Institute, Boston, Massachusetts
    Search for more papers by this author
  • Rosemary Costello R.N.,

    1. Head and Neck Oncology Program, Dana Farber Cancer Institute, Boston, Massachusetts
    2. Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts
    Search for more papers by this author
  • Maryann Case R.N.,

    1. Head and Neck Oncology Program, Dana Farber Cancer Institute, Boston, Massachusetts
    2. Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts
    Search for more papers by this author
  • Marshall R. Posner M.D.

    1. Head and Neck Oncology Program, Dana Farber Cancer Institute, Boston, Massachusetts
    2. Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts
    Search for more papers by this author

  • Data management support was provided by Aventis and Rhône-Poulenc Rorer.



The authors conducted a Phase I/II study in patients with a poor prognosis who had locally advanced squamous cell carcinoma of the head and neck (SCCHN) and who were treated initially with induction chemotherapy. Patients were treated with weekly docetaxel and concurrent daily fractionated radiation therapy to determine the maximum tolerated dose (MTD) of docetaxel and the efficacy of the regimen.


Twenty-two patients were enrolled, and 21 patients were treated. Eight patients had Stage III SCCHN, and 13 patients had Stage IV SCCHN without distant metastases and were treated first with 2-3 cycles of induction chemotherapy, which consisted of cisplatin plus 5-fluorouracil with or without leucovorin. Patients with a poor prognosis were identified as those who achieved a partial response to induction treatment, achieved a complete response with a positive biopsy, or were at high risk for developing recurrent disease. Patients were treated subsequently with concurrent, escalating doses of docetaxel (given weekly × 6) and once daily 200-centigray radiation fractions.


Three patients were treated with a weekly docetaxel dose of 20 mg/m2 without dose-limiting toxicity (DLT). Both patients who were treated at the next dose level of 30 mg/m2 experienced DLT. A dose of 25 mg/m2 was studied without DLT in the 16 patients who were treated, establishing this as the MTD. Sixty-seven percent of the patients are alive without disease at a median follow-up of 35 months (range, 12–59 months) after the initiation of chemoradiotherapy.


The MTD of weekly docetaxel with concurrent daily radiation therapy in the postinduction setting was 25 mg/m2. Disease free survival data from this study were good and indicated that this regimen was effective in the treatment of patients with SCCHN who had a poor prognosis. Cancer 2002;95:1472–81. © 2002 American Cancer Society.

DOI 10.1002/cncr.10873

Induction chemotherapy followed by radiation is an established, effective approach for achieving organ preservation and improving survival in patients with squamous cell carcinoma of the head and neck (SCCHN).1–6 Phase III studies have demonstrated the efficacy of combined chemotherapy and radiation for maintaining a functional larynx for patients with either a laryngeal or hypopharyngeal primary tumor without altering survival and for improving survival in patients with unresectable disease.1, 3–6 Patients who do not have a significant response to the first two cycles of chemotherapy frequently go directly to surgery or radiation therapy. A common finding in these trials is that patients who achieve a primary site clinical and/or pathologic complete response (CR) to three cycles of induction chemotherapy fare well in terms of local disease control. In contrast, patients who have a partial response (PR) or a positive biopsy at the primary site have not done well when they were treated with the standard postinduction local therapy of daily radiation. For example, in the Veterans Administration larynx study, there was an ≈ 70% (pathologically negative) versus ≈ 35% (pathologically positive) difference in 3-year disease free survival based on preirradiation biopsies.5 In managing the care of patients who are treated on induction chemotherapy regimens, it is important to ask is whether a more aggressive, local-regional therapy combination may improve the outcome for patients with high risk, poor prognosis disease who have not responded adequately to induction chemotherapy.

Concurrent chemoradiotherapy yields an increase in local-regional dose intensity relative to radiotherapy alone and is an effective method for improving response rates in patients with SCCHN. Numerous systemic agents have been employed for concurrent regimens, including 5-fluorouracil (5-FU), cisplatin, hydoxyurea, and bleomycin.7–13 Recently, there has been significant interest in the use of the taxanes (docetaxel and paclitaxel) in concurrent, combined-modality treatment programs for treating patients with solid malignancies, including SCCHN.14–24 The rationale for integrating the taxanes into combined treatment regimens has been based on 1) the proven efficacy of the taxanes both as single agents and as components of multiagent chemotherapy programs in the treatment patients with of SCCHN and 2) the in vitro data demonstrating the beneficial, interactive effect of concurrent taxane and radiation.25–30 Both of these factors have contributed to the use of combinations of taxanes and radiation for the treatment of patients with SCCHN in many clinical situations, including concurrent chemoradiotherapy as part of an initial treatment for patients with locally advanced disease.14–22 In addition to the factors that support the use of concurrent taxanes and radiation, patients in the postinduction chemotherapy setting may have an additional benefit. Laboratory data indicate that the combination of a taxane and radiation is effective in cells that have increased primary resistance to other drugs.31 This suggests that a taxane-radiation combination may have significant efficacy in the postinduction setting for patients with a poor prognosis who have not responded well to induction therapy. Because local-regional recurrence remains the main site of failure, the use of low-dose, weekly, concurrent drug treatment is an ideal schedule, because it maximizes drug-radiation interaction and increases local-regional dose intensity with limited systemic toxicity. The sequential treatment program described here takes advantage of the systemic activity of the high-dose induction treatment initially delivered and allows the clinician the flexibility to determine the appropriate extent of local-regional dose intensity based on the response to prior induction therapy and the patient's prognosis.2

We performed a Phase I/II trial examining the use of weekly docetaxel with concurrent, daily radiation for patients with SCCHN who had a poor prognosis after induction chemotherapy. The eligible population was defined as patients who received induction chemotherapy and had a clinical CR or PR with positive biopsies at the primary site, a primary site nonresponse, persistent lymph node disease, or an extremely poor prognosis based on pretreatment clinical factors. Radiation was delivered on a schedule using 200-centigray (cGy) daily fractions in combination with weekly docetaxel. Initially, the docetaxel dose was escalated to determine the maximum tolerated dose (MTD) of the drug, which was established at 25 mg/m2 per week for 6 doses. After establishing the MTD, 16 patients were treated at this dose level to obtain additional safety and efficacy data.


The protocol used was approved by the Institutional Review Board, and all patients went through an informed consent process prior to entry on the trial. Patients were evaluated and treatment recommendations were made at a multidisciplinary clinic that included medical oncologists, radiation oncologists, head and neck surgeons, and dentists. The patients who were included in this trial had biopsy-proven, locally advanced, Stage III or IV SCCHN. Prior to their diagnosis, patients had not received chemotherapy or radiation for SCCHN. Patients were required to receive one to three cycles of an induction chemotherapy regimen as initial treatment; the patients who actually enrolled received a minimum of two cycles of chemotherapy. Acceptable chemotherapy regimens consisted of cisplatin, 5-FU, and leucovorin (PFL), as described previously, or cisplatin and 5-FU, with a minimum cisplatin dose of 80 mg/m2 and a minimum 5-FU dose of 1000 mg/m2 per day for 4 days.32 Patients were offered enrollment in this study if they 1) achieved a PR or a nonresponse (NR) to induction chemotherapy; 2) achieved a clinical CR to induction chemotherapy but had a positive postinduction biopsy; or 3) had a tumor that indicated a very poor prognosis (< 30% 2-year survival) based on pretreatment clinical factors, in the judgment of the treating physicians. Prior taxane treatment was not allowed, and patients were required to start concurrent docetaxel-radiation therapy within 13 weeks of the start of their induction chemotherapy. Furthermore, patients were required to be age ≥ 18 years, have an Eastern Cooperative Oncology Group performance status ≥ 2, and have adequate nutritional intake. Patients were excluded if they required intravenous alimentation to maintain their nutritional status. A normal bilirubin level, levels of aspartate aminotransferase or alanine aminotransferase < 1.5 times the upper limit of normal, and an alkaline phosphatase level < 2.5 times the upper limit of normal were required for enrollment.

Required prestudy testing for entry included a complete history and physical examination with baseline evaluation of toxicity and symptoms: complete blood count with differential and platelet count; hepatic function testing; blood urea nitrogen, electrolytes, and creatinine; serum calcium, magnesium, total protein, albumin, glucose, and uric acid; urine analysis; 12-lead electrocardiogram; posteroanterior and lateral chest X-ray; computed tomographic scans or magnetic resonance images of the sites of disease; and serum or urine human chorionic gonadotropin for women of child-bearing potential. A TNM staging assessment based on physical examination, and radiologic studies were performed within 4 weeks of study entry. When they were performed, an examination under anesthesia and a biopsy were used to support staging determinations. A percutaneous endoscopic gastrostomy (PEG) tube was placed in all patients prior to chemoradiation.

Prescribed treatment consisted of daily radiation delivered at 2 gray (Gy) per day to a total dose of 68–72 Gy concurrent with weekly docetaxel. The initial dose level of docetaxel was 20 mg/m2, and the second dose level was 30 mg/m2. An additional, intermediate, dose level of 25 mg/m2 was added due to toxicity at 30 mg/m2. The drug was given as an intravenous infusion on Monday, Tuesday, or before radiation on Wednesday, so that patients received a minimum of three postchemotherapy radiation treatments in one week. Patients were premedicated with dexamethasone (20 mg intravenously or orally) 6 hours and 12 hours prior to infusion, diphenhydramine (50 mg intravenously), and ranitidine (50 mg intravenously) or cimetidine (300 mg intravenously). Dexamethasone (10 mg intravenously) was given after docetaxel administration.

All patients were treated with external beam radiation using an initial course that typically consisted of two lateral fields and a supraclavicular field. Fields sufficient to encompass all known disease were used at the discretion of the individual physician with the goal of delivering a total dose of 68–72 Gy. During radiation treatment, patients were seen weekly to assess toxicity, which was graded according to National Cancer Institute Common Toxicity Criteria. Patients had post-treatment follow-up at the multidisciplinary clinic, including physical examination, history, and endoscopic examination. Follow-up cross-sectional imaging studies were performed when they were indicated. The duration of follow-up and the time to failure were defined as the interval from the beginning of concurrent docetaxel-radiation treatment to the present and to the time of development of recurrent disease, respectively.

The decision regarding neck dissection was made based on response to induction chemotherapy. Patients undergoing a lymph node CR to induction chemotherapy were not required to undergo dissection, regardless of the initial N stage. If patients had a PR/NR for the neck after induction chemotherapy, then a neck dissection was scheduled for 4–6 weeks after the completion of radiation. Surgical treatment of the primary lesion was not a planned part of overall management.

A CR was defined as complete absence of the initial tumor on clinical examination. A PR was defined as a decrease > 50% in the product of the two greatest perpendicular dimensions of the lesion (either the primary tumor or a lymph node). An NR was defined as less than a PR and included stable disease (SD) or progressive disease (PD). The dose-limiting toxicity (DLT) that was relevant to this trial was acute radiation toxicity, which was defined as a treatment break > 14 days.


Patient Population

In total, 22 patients were enrolled in the protocol between November, 1995 and November, 1999, and 21 patients received the prescribed combined-modality treatment plan and were evaluable for response and toxicity evaluation. One patient agreed to enroll in the trial after induction chemotherapy but withdrew prior to initiating concurrent chemoradiotherapy and, thus, prior to treatment as prescribed by the protocol. This patient is noted here to indicate all patients who were enrolled formally in the trial; however, in our subsequent analysis of results, we included only the 21 patients who actually initiated the concurrent docetaxel plus radiation therapy. Table 1 describes the patients who were included in this study. There were 13 males and 8 females with a median age of 60 years (range, 36–76 years). The sites of primary disease were as follows: six oropharyngeal lesions, one nasopharyneal lesion, eight laryngeal/hypopharyngeal lesions, five lesions of the oral cavity, and one unknown primary lesion. Table 2 illustrates the tumor and lymph node status distribution as follows: Tx, 1 patient; T2, 4 patients; T3, 12 patients; T4, 4 patients; N0, 8 patients; N1, 2 patients; N2, 7 patients; and N3, 4 patients. Fifty-two percent of patients had N2 or N3 disease (11 of 21 patients). One patient had a prior supraglottic laryngectomy. All patients had American Joint Committee on Cancer Stage III tumors (8 patients) or Stage IV tumors (13 patients), and their pathology demonstrated squamous cell carcinoma or one of its histologic variants.

Table 1. Description of the Patients Enrolled in the Trial, Induction Chemotherapy Prior to Entry, and Distribution by Primary Tumor Site
Patient characteristicsValue
  1. PF: cisplatin and 5-fluorouracil; PFL: cisplatin, 5-fluorouracil, and leucovorin.

Total entered (evaluable)22 (21) patients
Male:female ratio13:8
Age (yrs) 
Prior PF:PFL17:4
No. of cycles 
 1–22 patients
 319 patients
Primary site 
 Oral cavity5
Table 2. T Classification and N Classification of Treated Patients at the Time of Initial Presentation, Prior to the Initiation of Induction Chemotherapy
N classificationT classification

According to the protocol, all patients had received previous treatment with induction chemotherapy, with most patients receiving the planned 3 cycles (19 of 21 patients) and the other patients receiving 2 cycles. The majority of patients had an induction regimen consisting of PF (16 patients) or PFL (4 patients). One patient was treated with an amifostine and PF-based regimen. The responses to induction chemotherapy, including separate evaluations of primary sites and regional lymph nodes, are detailed in Table 3. For all sites, including both primary and lymph node sites, 19% of patients achieved a CR, and 76% of patients achieved a PR to induction chemotherapy. Seventeen of 21 patients (81%) underwent primary site biopsies under anesthesia after induction chemotherapy at the time of examination: The results of these biopsies are described in Table 4. Two patients did not have a primary site, one who had the prior supraglottic laryngectomy, and one with an unknown primary. Two patients did not undergo a biopsy. Among the eight patients who achieved a CR at the primary site and who underwent a biopsy, three patients (38%) showed biopsy evidence of disease. All patients who achieved a primary site clinical PR or SD had positive biopsy results. Overall, 12 of 17 patients (71%) who were entered on this trial and underwent a primary site biopsy had pathologically positive findings at the primary site. Notably, 11 of 13 patients had less then a CR in regional lymph nodes. Only four patients achieved a CR to prior induction therapy at all sites; three patients achieved a clinical CR but had positive biopsy results at the primary site, and one patient had an unknown primary tumor (TxN3) and achieved a CR in the neck. This patient was not scheduled to undergo a neck dissection by our standard treatment algorithm; however, we believed that he had a > 70% risk with treatment using standard radiation doses and, for this reason, was included in the trial. Thus, all patients but one were classified with a poor prognosis based solely on their clinical response or biopsy results.

Table 3. Pretreatment Responses to Induction Chemotherapya
VariableResponse (%)
  • CR: complete response; PR: partial response; NR: no response; RR: response rate.

  • a

    Clinical responses of the patient population to induction chemotherapy, prior to receiving concurrent taxotere-radiation.

Primary site9/19 (47)9/19 (47)1/19 (5)18/19 (95)
Lymph nodes2/13 (15)10/13 (77)1/13 (7)12/13 (92)
Overall4/21 (19)16/21 (76)1/21 (5)20/21 (95)
Table 4. Pretreatment Primary Site Biopsiesa
VariablePrimary site response (%)
CRPRNRTotal pos
  • CR: complete response; PR: partial response; NR: non response; Total pos: total pathologically positive responses.

  • a

    Results of primary site biopsies after induction chemotherapy.

  • b

    Two of 19 primary sites were not biopsied (1 CR and 1 PR).

Positive biopsy3/8 (38)8/8 (100)1/1 (100)12/17 (71)b

Response and Survival

The responses to the entire course of treatment, including induction chemotherapy and concurrent docetaxel plus radiation therapy, are summarized in Table 5. One of the key findings was the increase in the number of patients who achieved a clinical CR from 4 of 21 patients (19%) to 12 of 21 patients (57%). Overall, only 10% of patients were classified with SD or PD, whereas 1 patient remained inevaluable. There were 13 patients with lymph node disease at the time of initial presentation, and 11 patients had a PR or NR to induction chemotherapy. After receiving concurrent docetaxel-radiation therapy, 7 of 11 patients underwent neck dissection; 5 of those 7 patients had pathologically positive findings; and 3 of those 5 patients failed at 7 months, 24 months, and 27 months (2 patients with distant disease). The remaining two patients had no evidence of disease at 49 months and 56 months. The two patients with pathologically negative results have remained disease free. Six of 13 patients with neck disease at the time of initial presentation did not undergo neck dissection. Four of these six patients achieved a PR in the neck and did not undergo neck dissection for the following reasons: 1) physician choice and the patient was alive at 24 months, 2) the patient developed distant metastases prior to surgery, 3) the patient had inoperable PD, and 4) the patient died of suicide. The other two patients who did not undergo neck dissection both had achieved a prechemoradiation CR in the neck and were alive at 14 months and 53 months.

Table 5. Response to Docetaxel plus Radiationa
VariableResponse (%)
  • CR: complete response; PR: partial response; NR: non response; UE: unevaluable; RR: response rate.

  • a

    Clinical responses of the patient population after concurrent taxotere plus radiation.

  • b

    One patient committed suicide.

Primary site14/19 (74)3/19 (16)1/19 (5)1/19 (5)1/17 (89)
Lymph nodes6/13 (46)6/13 (46)1/13 (7)12/13 (92)
Overall12/21 (57)6/21 (29)2/21 (10)1/21 (5)18/21 (86)

The median follow-up for the 21 patients who were treated on this protocol is 35 months, with a range of 12–59 months. Currently, a total of 14 of 21 patients are alive without disease. Figure 1 shows event free, disease free, and overall survival. These data demonstrate an event free survival rate of 65% at 2 years. A total of six patients have developed recurrent disease at this point, and three patients are alive with disease. Of the living patients, one patient had a regional recurrence and is alive at 32 months, one patient had a local recurrence and is alive at 28 months, and a third patient had persistent disease and is alive after undergoing salvage surgery and the presentation of lung metastases at 29 months. Three patients died of disease. One patient developed recurrent disease in his primary site (tongue) and neck 3 months after treatment and died at 15 months. A second patient developed lung metastases and died at 4 months. A third patient failed with distant metastases, was treated with systemic therapy, and died at 25 months. Of the three patients who died of disease, two patients were treated at the lowest dose level. A fourth patient died of unrelated causes (suicide) at 3 months. In addition, the patient who withdrew from the trial prior to initiating therapy died at 10 months. One patient was withdrawn from the study for reduced performance status after receiving 4 doses of docetaxel and is alive at 17 months.

Figure 1.

Disease free survival (DFS), overall survival (OAS), and event free survival (EFS) of the 21 patients with squamous cell carcinoma of the head and neck who were treated with concurrent docetaxel plus radiation therapy on the Phase I/II protocol.

Treatment-Related Toxicity

Three patients were treated at the initial docetaxel dose of 20 mg/m2 per week without DLT. At the next planned dose level of 30 mg/m2, both patients who were treated experienced significant, acute toxicity. The first patient, who had a TxN3 unknown primary tumor, required a 13-day treatment break and was hospitalized with laryngeal edema, Grade 3 mucositis, and skin reaction. The second patient had a T2N1 recurrent oral tongue carcinoma and was hospitalized for mucositis, dehydration, and skin reaction after four weekly doses of docetaxel. This resulted in a DLT treatment break of > 14 days. Although the first patient's treatment break technically did not meet our criteria for DLT (defined as > 14 days), it was felt that no more patients should be treated at this dose level based on the extent of toxicity that was observed in these two patients. Both patients who were treated at the 30 mg/m2 dose level are alive without disease, although each patient required a late, temporary tracheostomy. Based on the toxicity experienced by the two patients who were treated at the 30 mg/m2 dose level and the minimal toxicity seen at the 20 mg/m2 dose level, an additional dose level of 25 mg/m2 was added to the protocol to assess the MTD. No further DLT was seen for the initial three patients who were treated at the 25 mg/m2 dose level; thus, this dose was established as the MTD and was used for the Phase II portion of the study. A total of 17 patients were enrolled, and 16 patients were treated at this dose level.

Acute and late toxicities are summarized in Tables 6 and 7. The predominant acute toxicities were mucositis and skin irritation, as expected from a combination of radiation with an effective chemotherapeutic-sensitizing agent. Excessive local reaction led to the toxicities at the 30 mg/m2 dose level. Late toxicities also were marked in the 30 mg/m2 group, with both patients requiring a temporary tracheostomy and delayed PEG tube removal (at 8 months and > 50 months). The degree of toxicity was significantly better in the patients who were treated at the 25 mg/m2 dose level, although there were delayed PEG tube removals (6 of 16 patients had tubes for > 6 months, 3 patients had tubes removed by 12 months, and 1 patient had a tube for > 24 months) and requirements for temporary tracheostomies (3 of 16 patients). One patient treated at 25 mg/m2 will likely have a permanent tracheostomy and PEG. Only one of three patients at the 20 mg/m2 dose level is a long-term survivor, and that individual had no long-term toxicity.

Table 6. Acute Toxicitya
ToxicityNCI CTC
20 mg/m2 per week (n = 3 patients)25 mg/m2 per week (n = 16 patients)30 mg/m2 per week (n = 2 patients)
Grade 3Grade 4Grade 3Grade 4Grade 3Grade 4
  • NCI CTC: National Cancer Institute Common Toxicity Criteria; XRT: radiotherapy.

  • a

    Results for acute toxicity are presented based on docetaxel dose levels.

XRT breaks122
Duration (days)43, 1113, 17
Table 7. Late Toxicitya
Toxicity20 mg/m2 per week25 mg/m2 per week30 mg/m2 per weekTotal
  • a

    Results for late (long-term) toxicity are presented based on docetaxel dose levels.

  • b

    More than 6 months after the completion of treatment.

  • c

    Temporary (long term, disease free).

  • d

    Long term (>1 year).

No. of patients316221
Delayed PEG removalb1/36/162/29/21
Esophageal stenosisd0/11/160/21/21


In the current study, we determined the MTD of docetaxel delivered once per week with concurrent, daily radiotherapy. The patient population had locally advanced SCCHN, had been treated previously with induction therapy, and had a poor prognosis based on their response to chemotherapy. Overall, the results of treatment for this group of patients with a poor prognosis was excellent, with 14 of 21 patients (67%) alive with no evidence of disease at a median follow-up of 35 months. Our data show an actuarial disease free survival rate of 70% at 24 months (Fig. 1) for our patients with varied primary sites. Historic controls for this group were not readily available; however, for the group of pathologically positive patients in the Veterans Administration Larynx Trial, the disease free survival rate at 24 months was 40%.5 The sequential integration of surgery after radiation therapy for patients with residual lymph node disease proved effective in two of five patients with positive pathologies. Our sequential regimen appears to be effective and offers a more aggressive local-regional treatment approach for selected patients who respond poorly to induction chemotherapy. In patients who are at high risk for recurrence, this can be considered in lieu of standard daily radiation. In our dose-escalation study, unacceptable, acute, local toxicity occurred at a weekly dose of 30 mg/m2 per week. At the MTD of 25 mg/m2 per week, acute toxicity was formidable but acceptable, and recovery was prolonged in a significant fraction of patients.

Induction chemotherapy has an established role of organ preservation in the treatment of patients with laryngeal and hypopharyngeal primary tumors and has been studied as a route to achieve organ preservation in other head and neck sites.1–6 The most frequently used radiation schedule after induction therapy has been daily treatment at a dose of 1.8–2.0 Gy per day up to a total dose of 65–70 Gy, although other schedules have been used. This approach has yielded acceptable results overall in terms of functional preservation, although the outcome for patients who do not respond well to induction treatment has not been good. In both PFL and docetaxel plus PFL regimens, persistent lymph node disease after induction therapy was associated with a poor outcome.32, 33 Patients who had a positive primary site biopsy after treatment in the Veterans Administration Larynx Trial (12% of patients who achieved a CR and 55% of patients who achieved a PR) had a 4-year disease free, local-regional survival rate of ≈ 35%, compared with an ≈ 80% survival rate for patients who had negative biopsies (P < 0.0001).5 The prescribed radiation regimen in this trial was daily fractions up to a total dose of 66–72 Gy. In the overall management of patients who were treated on induction protocols, these data raise the question of how to improve outcome for patients with a poor prognosis who have historically done poorly with standard radiation therapy. One approach is to minimize the number of patients with a poor prognosis by employing a more intensive induction treatment with the goal of increasing the pathologic CR rate. Along with other groups, we have studied this by incorporating additional drugs into the standard induction regimen of platinum and 5-FU.2, 33 The inclusion of more cytotoxic agents or higher doses in the induction regimen may prove to be a more effective approach. However, dose intensification is not applicable universally, because many intensive protocols add significant toxicity, thus limiting the potential patient population, and it may delay the start of potentially curative radiation therapy. Another alternative is to use standard induction regimens and put the emphasis on increasing the efficacy of postinduction local therapy for high-risk patients. This is the approach we have pursued in the current study by developing a postinduction combination of daily radiation and concurrent weekly docetaxel. Others have shown an improvement in local control and overall survival when concurrent chemotherapy is added to a given radiation schedule.11, 12

The concurrent use of the taxanes (docetaxel and paclitaxel) with radiation has been employed extensively for the treatment of patients with many types of solid tumors, particularly nonsmall cell lung and head and neck malignancies. This concomitant use is based on efficacy of the single agent taxane as a cytotoxic treatment for those classes of tumors and, more importantly, extensive in vitro data demonstrating a therapeutic benefit for its use combined with radiation. The initial hypothesis for the radiosensitizing effect of the taxanes was based on induced cell cycle changes.25, 26 The microtubule modification initiated by the taxanes caused cells to accumulate in the radiosensitive G2/M phase of the cell cycle. This mechanism contributes to the overall efficacy of taxane-radiation combinations; although, clearly, there are other relevant pathways and mechanisms. Further in vitro and in vivo data have demonstrated roles for the p53 pathways and apoptotic mechanism as contributing factors in the therapeutic benefit of combining a taxane with concurrent radiation.34, 35 Additional in vivo studies also have shown a clear differential between the effect of this combination on malignant tissue compared with normal tissue.36 This evidence of an increase in therapeutic ratio is a further rationale for continued clinical use of the agents. In the specific clinical situation represented in this study, patients with chemotherapy resistant disease, the laboratory gives yet another reason for pursuing this combination. In vitro studies of paclitaxel plus radiation showed enhanced radiosensitivity in cell lines that demonstrated drug resistance.31 These experimental data support the potential for a significant role of concurrent chemoradiotherapy with taxanes in patients with poor responses to induction chemotherapy.

There are various theories and approaches with regard to the optimal combination of taxanes and radiation. However, the use of frequent drug dosing (daily or weekly) offers the longest duration of exposure and potential for maximizing the drug-radiation interaction. In addition, there are accumulating data that a weekly schedule of taxanes may be the most effective cytotoxic dosing.37 For both of these reasons, weekly docetaxel has been the most widely investigated schedule in combination with radiation. Every 3-week dosing of the drug is a more standard regimen when using the drug as a cytotoxic agent. However, if this approach is employed in a concurrent treatment regimen, then it is being used on a distinctly different conceptual basis compared with the more frequent weekly or daily dosing regimens.14 Although there are increased plasma levels resulting from the higher doses given every 3 weeks, only a small fraction of the radiation treatment will be sensitized. In that case, the combined-modality approach relies more on the drug and radiation as noncross-resistant therapies than on the radiosensitizing interaction. The patients who were treated in the current study were treated previously with cytotoxic doses of chemotherapy as part of the induction regimen; therefore, we focused on using a schedule that would maximize a radiosensitizing schedule in an effort to increase historically low rates of local control.

For the treatment of patients with solid tumors, the use of concurrent docetaxel with radiation has been investigated in patients with nonsmall cell lung carcinoma (NSCLC) and SCCHN. Data from two dose escalation trials in patients with NSCLC suggest that a 20–30 mg/m2 dose of docetaxel would be the MTD for weekly therapy.38, 39 Although a number of trials have examined the use of docetaxel as part of multiagent treatment regimens for the treatment of patients with SCCHN at the time of initial presentation or for recurrent disease, only two trials have examined concurrent docetaxel with radiation.21, 42 One study examined concurrent weekly docetaxel plus irinotecan with radiation to a total dose of 66–70 Gy.42 The MTD consisted of 20 mg/m2 for docetaxel and 55 mg/m2 for irinotecan. The data suggest that the MTD of docetaxel given as a single agent concurrent with radiation would be at least 20 mg/m2. In contrast with this result, a study of docetaxel as a single agent in patients with Stage IV SCCHN had multiple DLTs (two Grade 4 skin reactions, two Grade 4 pulmonary toxicities, and one Grade 3 thrombocytopenia among 6 patients) at the initial weekly dose of 15 mg/m2.21 The reason for the failure of that trial even to complete the first dose level is not clear.

In addition to the docetaxel studies, there are also data for concurrent, single-agent paclitaxel and radiation in patients with SCCHN. Our group has published a study using paclitaxel every 3 weeks at a dose of 100 mg/m2 with daily radiation of 60–72 Gy in a heterogeneous patient population, most of whom were treated initially with induction chemotherapy.14 Overall responses were good, and the local toxicity was managed effectively with aggressive use of PEG tubes. A trial of weekly paclitaxel with concurrent daily radiation to 60–70 Gy demonstrated an MTD of 30 mg/m2, which was similar in dose intensity (90 mg/m2 every 3 weeks) to our regimen with a DLT of mucositis.20 A completed Phase I study using a 96-hour continuous infusion delivery of paclitaxel given during Weeks 1 and 5 of radiation with a modified, concomitant boost radiation schedule had an MTD of 100 mg/m2 and DLTs of skin toxicity, mucositis, and febrile neutropenia.19 Eight patients achieved a CR, 4 patients achieved a PR, and 1 patient had PD among 13 patients who were treated, with all patients experiencing reversible Grade 3 toxicity. Plasswilm et al. employed a unique, split-course/twice-daily radiation schedule with concurrent paclitaxel delivered during Weeks 1 and 5 at a dose of 30 mg/m2 per day for a total of 5 days.18 Ten of 12 patients completed that prescribed therapy, with 8 patients achieving a CR and 2 patients achieving a PR. There also are two ongoing trials, one study examining continuous infusion paclitaxel, which has not yet reached DLT, and one study that will add growth factor support to determine whether paclitaxel doses given every 3 weeks can be increased.15, 22

The acute toxicities experienced by our patients with advanced-stage SCCHN with this regimen were significant but were well managed with the aggressive use of PEG tubes. Our local DLT, at 30 mg/m2, was consistent with results reported by other groups in which DLTs typically were local rather than hematologic. The long-term toxicity experienced by these patients is a cause of potential concern. There was an extended period before removal of the PEG tube for a number of patients, as expected in an aggressive, dose-intensive therapy. The multiple temporary tracheostomies also are an area of concern. Both patients who were treated at the 30 mg/m2 level, both of whom had DLTs secondary to acute toxicity, also experienced long-term toxicity in the form of tracheostomy and long-term PEG placement. These data clearly indicate that there is a risk of severe acute and late toxicity that must be monitored closely. It is noteworthy that the delayed esophageal stenosis experienced by three patients has been treated successfully with dilitation.

The approach described herein has shown significant efficacy in treating a patient population with a poor prognosis. The approach merits consideration in treating these patients after induction chemotherapy and also may be considered as an initial treatment regimen for patients with locally advanced SCCHN. The local therapy was based on a standard radiation schedule; however, since the initiation of our study, the results of Radiation Therapy Oncology Group Trial 90-03 have been published and clearly identified the most effective radiation schedules for treating patients with advanced SCCHN.43 That study showed that radiation treatment using 120 cGy twice daily and the concomitant boost technique were more effective than daily radiation or 1.6 Gy twice daily with a break. For our next trial, we are integrating one of the more effective schedules, the concomitant boost technique, with concurrent docetaxel. We are investigating a concurrent once-a-day radiation and weekly docetaxel combination for 4 weeks followed by radiation twice daily without concurrent chemotherapy. The treatment regimen will be given with 180-cGy fractions daily and completed with twice-daily radiation using a 180-cGy/150-cGy combination, according to the concomitant boost technique. Long-term toxicity in this new trial should be less compared with long-term toxicity in the current study, because the radiation fraction size is lower (180 cGy and 150 cGy vs. 200 cGy), and it has been shown that fraction size is a key parameter in determining late effects of radiation. In addition, only four doses of drug will be given overall, compared with six doses in the current trial. Our goals are to increase the local and disease free survival of our population and to decrease local toxicity by integrating what has been proven to be an optimal radiation schedule with the concurrent, weekly docetaxel regimen that, as demonstrated in the current study, has significant efficacy.