A Phase II Trial of Cetuximab and Docetaxel With Radiation for Head and Neck Cancer

Vivek N. Patel, MD

Objectives:

This study evaluates low-dose radiotherapy in combination with docetaxel and cetuximab in patients with recurrent, unresectable squamous cell carcinoma of the head and neck initially treated with definitive chemoradiation therapy.

Methods:

In this phase II study, cetuximab 400 mg/m2 was given in the usual loading dose only on day 1 of week 1 and then, subsequently, in 250 mg/m2 weekly doses on day 1 of each following week (weeks 2-7). Docetaxel 20 mg/m2 was administered weekly on day 1 from weeks 2-7. Radiotherapy consisted of 0.5 Gy/fraction twice a day at least 6 hours apart on days 2 and 3 of weeks 2-7 for a total dose of 12 Gy.

Results:

Nine patients were screened, 5 patients were enrolled, and 4 patients were treated between October 2013 and February 2015. After completion of salvage treatment per protocol, at 1-month follow-up, partial response was seen in 2 patients (50%), stable disease in 1 patient (25%), and progressive disease in 1 patient (25%). By 6 months, all patients had progressive disease, 1 patient (25%) completed a full course of re-irradiation, and 1 patient (25%) received further chemotherapy for progression. Treatment-related morbidity was limited to difuse grade 1 skin toxicity reported in all patients. The Data Safety and Monitoring Board suggested closing this protocol as clinical response approached the prede ned futility endpoint for complete response.

Conclusions:

Low-dose radiotherapy in addition to docetaxel and cetuximab was well tolerated, although treatment effects were transient. More durable treatment options may exist for definitive treatment; however, this protocol may offer palliation to select patients.

INTRODUCTION

Local/regional failure after definitive radiotherapy (RT) remains a significant problem, particularly in advanced-stage head and neck cancer.^1-3Approximately 50% to 60% of patients who fail initial definitive treatment will die as a direct consequence of locally or regionally recurrent disease.^1,2,4Surgical salvage is sometimes successful, but it is not always feasible due to disease extent, location, or patient comorbidities.⁵For these patients, systemic therapy, reirradiation, and supportive care may be the only remaining options.^6-9

Historically, prior to when reirradiation was not yet commonly used, systemic therapy alone was considered the remaining palliative treatment option for inoperable patients with recurrent squamous cell carcinoma of the head and neck (SCCHN). Use of platinum-based chemotherapy has been associated with response rates of less than 50% and a median overall survival (OS) of 6 to 8 months.^7,10More recently, cetuximab therapy has demonstrated improved response rates and survival in platinum-refractory SCCHN compared with historical controls.^11,12In a study of 442 patients randomized to receive a platinum-based chemotherapy and fluorouracil with or without cetuximab, the addition of cetuximab significantly prolonged the median OS to 10.1 versus 7.4 months in the control arm.¹³Furthermore, the same authors reviewed 3 separate phase II trials in patients with recurrent or metastatic SCCHN who progressed on platinum-based therapies, and found that the addition of cetuximab led to additional improvements in response rates from 10% to 13%.¹⁴The authors concluded that cetuximab may prolong survival in patients with SCCHN who fail first-line therapy.

Reirradiation of the head and neck may be fraught with difficulties in relation to concerns regarding normal tissue tolerance and, sometimes, tumor radioresistance.¹⁵More recently, preclinical and clinical data have shown that reirradiation is feasible and may provide improved disease control, translating into superior survival in subsets of patients; that is, however, in the setting of a potentially high toxicity profile.^16,17Various types of RT delivery, such as con- ventionally fractionated therapy, hyperfractionation, brachytherapy, and stereotactic body RT, have been attempted with and without concurrent chemotherapy in this group of patients. In appropriately selected patients who are able to undergo reirradiation, there is usually an improved disease response compared with systemic therapy alone; however, there also continues to be a notable risk of treatment-related morbidity with such combination therapy.^18-20

In contrast to the paucity of treatment options for patients in the loco-regionally recurrent and unresectable setting, there have been numerous developments in the primary treatment of SCCHN. Low-dose fractionated RT (LDFRT) is a novel approach that has been utilized in the treatment of loco-regionally very advanced SCCHN as a means of decreasing the toxicity of reirradiation. LDFRT resulted in more effective cell-kill than was predicted by the linear quadratic model in preclinical studies, possibly through both the hyperradiation sensitivity (HRS) phenomenon and/or through potentiating the effect of chemotherapy.^21-24Early work by groups such as Joiner and colleagues showed an initial phase of HRS in head and neck cell lines that were exposed to doses less than 1 Gy, specifically around 0.5 Gy.²³In this low-dose region, there exists a paradoxical increase in tumor cell kill compared with higher doses. It is hypothesized that HRS does not cause the initiation of DNA repair mechanisms that would otherwise arrest these cells for repair in the second gap (G2) phase. As such, these damaged cells ultimately undergo cellular death.^23,24

The use of LDFRT along with chemotherapy has also demonstrated a chemo-potentiating role. In vitro data initially revealed that LDFRT potentiates the effects of the paclitaxel in both wild type and mutant p53 head and neck tumor cell lines.²²Furthermore, LDFRT in both cell lines caused an up-regulation in the pro-apoptotic protein, BAX, and a down-regulation in the anti-apoptotic proteins BCL-2 and NF-κB. The combined effects of LDFRT and docetaxel therapy on the growth of SCCHN were also evaluated in the nude mouse model to elicit the HRS phenomenon present in G2/mitosis (M) cell cycle.²¹Combination therapy showed a significant improvement in both tumor regression and local control. Additionally, molecular analysis of resected tumor specimens demonstrated that BAX levels were elevated with concomitant increase in cytochrome c oxidase release to the cytosol in the LDFRT and docetaxel group. These findings strongly suggest that LDFRT can be used in combination with docetaxel to potentiate the effects of docetaxel on tumor regression through an apoptotic mode of death. As such, the G2/M cell cycle arrest by docetaxel appears to be an important component of the enhanced apoptotic effect of LDFRT + docetaxel combined treatment.

The aforementioned in vitro and in vivo studies also led to a phase II clinical trial. Gleason and colleagues evaluated 39 patients with loco-regionally advanced SCCHN who were treated with induction paclitaxel, carboplatin, and LDFRT every 21 days for 2 cycles.²⁵RT consisted of 80 centigray (cGy) twice a day on days 1 and 2 to a total dose of 640 cGy. Initial results were promising and showed a response rate of 90% at the primary disease site and 69% at the nodal site.²⁶Five-year follow-up was similar to historical controls, showing an OS of 62% and progression-free survival (PFS) of 58%. In this study, treatment-related toxicity was minimal.

Our novel approach to treat recurrent SCCHN was based on the following 5 hypotheses, extrapolated from the aforementioned data: 1) LDFRT would be utilized to sensitize chemotherapy, 2) low-dose HRS response would be significantly enhanced in docetaxel-induced G2/M cell cycle arrest, 3) LDFRT would render an enhanced BAX-activation—mediated mode of apoptotic cell death, 4) cetuximab would help arrest the cells in the inactive, noncycling G1/ G0 state leading to a p21-mediated mode of cell death; and 5) the toxicity profile would be expected to be minimal. The purpose of this study is to describe our initial exploratory experience with LD- FRT in addition to docetaxel and cetuximab in the setting of recurrent SCCHN.

Materials and Methods

Study and patient characteristics/eligibility criteria

This study was Institutional Review Board—approved, Investigational New Drug–exempt, and Health Insurance Portability and Accountability Act–compliant. In addition to having loco-regional recurrence in the head and neck, all patients fulfilled all the following inclusion criteria: 1) Eastern Cooperative Oncology Group performance status of 0-2, 2) histo-pathologically confirmed recurrence of SCCHN, 3) not candidates for surgical salvage treatment due to unresectability, and 4) a period of at least 6 months from the completion of prior RT had passed. Exclusion criteria included distant metasta- ses outside the head and neck, primary disease in the nasopharynx or salivary glands, other invasive malignancies, or medical illnesses impairing ability to receive treatment.

Enrollment and study size

The plan was to enroll 25 evaluable patients and follow them for a minimum of 2 additional years. The primary objective was to show an improvement of response rate, from 40% with chemotherapy alone to 70% with the addition of LDFRT. Considering the possibility that 5% of study-eligible patients will not be evaluable, we aimed to enroll 27 patients to ensure an analysis set of 25 patients who would be evaluable for efficacy. This sample size would provide reasonable precision in estimating the overall response rate (ORR) as preliminary evidence that the proposed docetaxel-cetuximab-LDFRT treatment is efficacious in this population. ORR is defined as rate of patients achieving complete response (CR) or partial response (PR). Secondary objectives consisted of a safety profile along with an estimate of PFS and OS.

Chemotherapy and radiation

Cetuximab 400 mg/m2 IV was administered as a loading dose 1 week prior to RT and then cetuximab 250 mg/m2 IV was given weekly on day 1 of weeks 2 to 7. Docetaxel 20 mg/m2 IV was given weekly on day 1 of week 2 to 7. LDFRT consisted of 0.5 Gy/ fraction given twice a day via intensity-modulated radiotherapy at least 6 hours apart on days 2 and 3 of weeks 2 to 7 for a total dose of 12 Gy. Total therapy lasted for a period of 7 weeks. The Figure reviews the treatment schema.

Evaluations during treatment

Patients were assessed weekly in the Department of Radiation Oncology, Sylvester Comprehensive Cancer Center, University of Miami Health System. Every week, complete blood counts and chemistries were evaluated to evaluate hematologic toxicity. The National Cancer Institute’s Common Terminology Criteria for Adverse Events, version 4.0, was utilized for documentation of toxicity.

Posttreatment evaluation

Follow-up visits consisting of history and physical examination were performed at 1, 3, 6, and 9 months. Four weeks after treatment, a computed tomography or magnetic resonance imaging scan of the head and neck was obtained. Response was evaluated by the treating physician defined as follows: CR was defined as disappearance of the target lesions; PR was defined as at least a 30% decrease in size of the lesion; progressive disease (PD) was defined as at least a 20% increase in the size of the lesion; and stable disease (SD) was defined as insuficient changes to qualify for the aforementioned endpoints of PR or PD. Patients who had less than a CR were recommended for further treatment with RT or chemotherapy at the discretion of the treating physician.

Early stopping guidelines

Proposals were set in place for the Data Safety Monitoring Committee (DSMC) at the Sylvester Comprehensive Cancer Center to stop the trial based on response or toxicity data. These guidelines were developed using Bayesian methods that could be applied at any phase of enrollment without advance specification of the number of interim analyses to be performed. Under these methods, a prior probability is assigned to possible values for response and toxicity data. As data on treated patients become available, the probability distributions are revised and the resulting posterior probability becomes the basis for recommending either continuation or termination of the study.

RESULTS

Patients

Between October 2013 and February 2015, 9 patients were screened for this institutional study to receive LDFRT, cetuximab, and docetaxel. Four patients were found to be ineligible for the study as 1 patient was found upon further work-up to have metastatic disease, 1 patient had a poor performance status, 1 patient elected hospice, and 1 patient died prior to enrollment. Five patients were enrolled in this study; however, 4 patients were ultimately treated because 1 patient was found to have a poor performance status and was not recommended for further treatment at the discretion of the treating physician. Table 1 lists the charac- teristics of the patients included in this study. All patients were male. Primary disease sites were the larynx in 2 patients, floor of mouth in 1 patient, and the tonsillar fossa in 1 patient. Only 1 patient received surgery as treatment for his primary tumor, followed by RT consisting of 66 Gy, whereas 3 patients received definitive chemoradiation consisting of cisplatin 100 mg/m2 every 3 weeks and RT dose of 70 Gy to their primary tumor. Time from end date of definitive primary treatment to therapy for recurrent disease ranged from 13 to 26 months. Four patients completed the reirradiation protocol and were able to tolerate treatment.

Acute toxicity

All the patients had grade 1 fatigue attributed to both chemotherapy and radiotherapy treatment. All patients had grade 1 rash ascribed to chemotherapy. There were no grade 2 or higher toxicities.

Response

At 1-month follow up, PR was seen in 2 patients (50%), SD in 1 patient (25%), and PD in 1 patient (25%). By 6 months, all patients had PD. One patient (25%) underwent a second course of reirradiation consisting of 58 Gy in 2 Gy/fraction as salvage treatment at progression, and 1 patient (25%) underwent chemo- therapy. Table 2 summarizes treatment response. Given that the accumulating data suggested that the ORR was no better than 40%, the DSMC recommended early discontinuation due to lack of efficacy as previously outlined.

DISCUSSION

To our knowledge, this is the first study to evaluate the use of LDFRT with chemotherapy in patients with recurrent SCCHN. The primary endpoint for assessing efficacy was ORR. The secondary objectives were treatment-related acute and long-term toxicities and estimate of PFS and OS. In this exploratory analysis, we found an initial 75% rate of stable or responsive disease. The treatment effect was transient, with all patients experiencing PD at 6 months. The treatment was tolerated well with all patients experiencing grade 1 fatigue and rash and no reported grade 2 or higher toxicities. The DSMC suggested closing this protocol early as results were approaching the predefined futility endpoint.

In the setting of nonsurgical treatment options for patients with recurrent SCCHN, there is a paucity of randomized data evaluating the comparative benefit of treatment options. Provider treatment biases as demonstrated in part by selection of healthier patients into surgical series, as well as the heterogeneity present in the remaining patient population, may account for this dearth of comparative clinical data. Consequently, national phase III studies such as Radiation Therapy Oncology Group 0421, a randomized trial designed to evaluate concurrent reirradiation and chemotherapy versus chemotherapy alone in patients with recurrent head and neck cancer, were closed early due to inadequate patient accrual.

Even after overcoming the challenge of enrolling patients, there are numerous difculties in maintaining a clinical trial within this select, chronically ill, and unstable population. Nine patients were originally screened, but only 4 were treated due to rapidly deteriorating functional status of the excluded patients. Furthermore, this trend of incomplete clinical trial data continues as this trial was closed early.

Prior to our protocol design, a study evaluating LDFRT, paclitaxel, and carboplatin in patients with primary locally advanced SCCHN was initially reported by Arnold and colleagues.²⁶In their study, the response rate was initially high at the primary site and nodal site. Our study was in part motivated by these developments; however, more recently these originally promising results were published as a 5-year follow-up with the mature data matching historic data with respect to disease-free survival and OS.²⁵In the recurrent SCCHN setting, the treatment effect with LDFRT and chemotherapy appears to be more transient, as our experience demonstrates. This may be in part due to a lower threshold for DNA repair due to previous treatment, or even an inherent and selected radio- resistance in cells surviving a previous full course of radiotherapy.

Early work with LDFRT unveiled the concept of HRS, in which low doses of radiation paradoxically increased tumor cell kill as compared with conventional doses.²³Hypothetically, HRS-related damage is below the threshold of DNA repair initiation that would otherwise be recognized (eg, double-strand damage that occurs after conventional doses). As such, the cell undergoes an apoptotic death.²⁵Depending on the study, this area of HRS is described as being in the <1 Gy range, usually described broadly from 0.1 to 0.8 Gy region of the cell survival curve.^22-24Notably, this area of HRS is swiftly followed by an area of increased radioresistance (IRR). IRR is thought to represent repair mechanism initiation after a dose-active increase in serine/threonine kinase activity and G2 checkpoint function.²⁷As such, radiation-induced damage triggers activation of repair mechanisms in the region of IRR. Some studies have described this area of IRR being present in the 0.3 to 1 Gy range.24,27 As such, it is plausible that 0.5 Gy may be too high of a dose to take advantage of HRS, and may even, conversely, fall into an area of IRR.

This is also the first study to use cetuximab and docetaxel in conjunction with LDFRT in patients with recurrent SCCHN. We based the use of cetuximab on the favorable response seen in patients receiving cetuximab in the recurrent setting, while docetaxel was chosen due to the potentiating effect of LDFRT based upon the aforementioned in vitro and in vivo clinical data.^14,22,25,28Although the potentiating effect of docetaxel is well described, there could be more of a potentiation role with other chemotherapeutic agents. It appears that LDFRT and both chemotherapeutic agents were very well tolerated compared with published experiences of multidrug chemotherapy regimens alone.^10,11,29

From our single institutional experience, patients with loco-regionally recurrent unresectable SCCHN should be treated with more established RT techniques, as they may not optimally benefit from this protocol.^2,8In patients with more limited options, as those presenting with metastatic disease, LDFRT with cetuximab and docetaxel may offer transient palliation with minimal morbidity.

CONCLUSIONS

Although there have been numerous improvements in the primary treatment of SCCHN, treatment of recurrent disease in inoperable patients is challenging. From our single-institutional experience, low-dose RT in addition to cetuximab and docetaxel is well tolerated in patients with recurrent SCCHN. Additional long-term treatment options may exist for patients with recurrent SCCHN; however, for a select group of patients, LDFRT and chemotherapy may offer transient palliation in the setting of low morbidity.