Chemotherapy Head and Neck Cancer

Nicholas W. Choong

Everett E. Vokes

Squamous cell carcinoma (SCCHN) is the most common histologic type of cancer to involve the upper aerodigestive tract. The pathogenesis of SCCHN is strongly correlated with cigarette smoking and alcohol use.¹ The human papilloma virus (HPV), in particular HPV type 16, is implicated in the pathogenesis of SCCHN in the tonsil and base of tongue, and associated with an increased incidence in Western countries in the last decade.² Epstein-Barr virus (EBV) is implicated in the pathogenesis of nasopharyngeal carcinoma (NPC), which is common in endemic areas such as East Asia and North Africa.³

Anatomically, SCCHN is a heterogenous disease because various sites with different functions in the head and neck can be affected. However, because of the relatively low incidence of cancer in each anatomic site, the similarity in their etiology, clinical history, and therapy, most clinical trials in SCCHN have grouped them together up until recently. Given our increasing understanding of the distinct epidemiologic and prognostic picture of virus-associated head and neck cancer (HPV and EBV), future trials will likely be conducted in a site-specific manner.

The TNM staging system used for SCCHN integrates all clinically available information, including that obtained by endoscopy and radiographic evaluation.⁴ NPC is staged by the Tumor-Node-Metastasis (TNM) staging system, and less commonly, with the Ho staging.⁵ Approximately one-third of patients with SCCHN will present with early-stage disease (stages I and II), whereas two-thirds will present with locoregionally advanced SCCHN (LA-SCCHN). Distant disease at the time of presentation occurs in <5% of patients and the most common metastatic sites are lungs, bone, and liver.⁶

GENERAL PRINCIPLES OF TREATMENT

For treatment considerations, SCCHN can be divided into several general stages:

Early-stage disease—stages I and II.
LA disease—stages III and IV (MO).
Recurrent and/or metastatic disease.

Radiation therapy (RT) and/or surgery are the curative treatments for SCCHN. The choice of modality depends upon the site, stage, comorbid conditions, and available expertise in the treatment modalities. For early-stage disease, the intention of treatment is cure. Patients may be treated with surgery or RT, which have similar local control and survival rates, but local recurrences following RT could be salvaged with surgery.⁷^,⁸ The long-term survival rates for early-stage disease ranges between 70% and 90%.

Patients with LA-SCCHN (stage III/IV, MO) are treated with curative intent. LA-SCCHN can be further subdivided into:

Resectable LA-SCCHN.
Unresectable LA-SCCHN.
LA-SCCHN with added goals of organ preservation.

Patients with resectable LA-SCCHN are managed with surgical resection followed by adjuvant RT. The addition of concurrent chemotherapy to adjuvant RT resulted in significant improvement in survival and can now be considered the standard of care for many patients.⁹^,¹⁰ For patients with unresectable LA-SCCHN, concurrent chemoradiotherapy remains the current standard. In patients who have surgically resectable LA-SCCHN but with the possibility that a resection will result in severe morbidity, a neoadjuvant chemotherapy approach has previously been utilized with a goal toward organ preservation and can be considered as another alternative to initial surgery.¹¹

Only a minority of patients with SCCHN present with metastatic disease. More often, patients develop recurrent disease or distant metastasis following definitive local therapy.¹² Patients with recurrent and/or metastatic disease are treated with palliative intent with systemic chemotherapy.

CHEMOTHERAPY FOR NEWLY DIAGNOSED LOCOREGIONALLY ADVANCED DISEASE

Resectable Locoregionally Advanced SCCHN

Postoperative (adjuvant) RT should be considered for patients who are at high risk for local failure after surgery. Although no randomized trials have addressed the efficacy of postoperative adjuvant RT, retrospective reviews support its use.¹³^,¹⁴ Patients with extracapsular extension and/or positive surgical margins attained 5-year locoregional control rate of 59% with postoperative RT compared with 31% with surgery alone.¹³ Matched-pair analysis of patients treated with a neck dissection and adjuvant RT compared with surgery alone showed higher recurrence rates in patients who did not receive adjuvant radiotherapy (RT). This is especially evident in patients with high T stage, close margins (<5 mm), positive margins, extracapsular extension, positive nodes, invasion of the soft tissues of the neck, vascular invasion, perineural invasion, and >5 mm of subglottic invasion.¹⁴

The Intergroup 0034 trial was a randomized study evaluating three cycles of cisplatin/5-fluorouracil (5-FU) after completion of adjuvant RT compared with adjuvant RT alone, in patients with completely resected SCCHN.¹⁵ This study found
no difference in survival, disease-free survival, or locoregional failure rates between the two groups. Only the incidence of distant metastasis was reduced by the addition of chemotherapy (15% vs. 23%).

Initial evidence supporting concurrent chemotherapy and adjuvant RT was shown in a trial that specifically treated patients who had extracapsular lymph node extension. Cisplatin (50 mg weekly) delivered with RT improved overall and disease-free survival compared with RT alone (5-year overall survival 36% vs. 13%).¹⁶

This leads to two randomized trials, conducted by the European Organization for Research and Treatment of Cancer (EORTC)¹⁷ and the Radiation Therapy Oncology Group (RTOG).¹⁸ Both trials administered adjuvant cisplatin 100 mg per m² on days 1, 22, and 43 of RT to patients with completely resected SCCHN who had high risk features—extracapsular lymph node tumor extension and/or microscopically involved surgical margins. The EORTC trial reported a beneficial effect on overall survival and local control with adjuvant chemoradiotherapy. In the RTOG trial, despite having an improvement in disease-free survival in the adjuvant chemoradiotherapy arm, there was only a trend toward improved local control and overall survival rates. Combined analysis of the two trials showed an advantage in locoregional control and survival for patients receiving adjuvant chemoradiotherapy in the setting of poor-risk features, that is, extracapsular invasion and/or positive resection margins.¹⁹

The German ARO 96-3 study utilized different doses of RT based on patient risk factors and delivered concurrent cisplatin and 5-FU with RT.²⁰ The locoregional control rate and disease-free survival in the adjuvant chemoradiotherapy arm were significantly better than RT alone. Taken together, these trials provide evidence that adjuvant chemoradiotherapy with a cisplatin-based regimen improves locoregional control rates and disease-free survival, and improvement in overall survival appears very likely.

Weekly delivery of cisplatin (50 mg) with adjuvant RT in patients with LA-SCCHN who have extracapsular lymph node tumor extension was evaluated in a small randomized study.¹⁶ The concurrent cisplatin and adjuvant RT arm attained better locoregional control and survival compared to adjuvant RT alone.

RTOG 0234 evaluated nonplatinum and cetuximab-based chemotherapy in the adjuvant setting in a randomized phase II study.²¹ The study delivered RT with either cisplatin-cetixumab or docetaxel-cetuximab for patients with positive margins and/or two or more nodal metastases or extracapsular nodal extension. With a median follow-up for surviving patients of 2.5 years, 2-year overall survival rates were 69% and 79% for cisplatin-cetixumab and docetaxel-cetuximab arms, respectively. The improvement with docetaxel-cetuximab arm appears to be mostly because of distant control with 2-year distant metastasis rates 13% versus 26%.

Unresectable Locoregionally Advanced SCCHN

Concurrent Chemoradiotherapy

Concurrent delivery of chemotherapy with RT, theoretically enhances the local antitumor activity of RT and combined with its systemic activity may eradicate possible micrometastases outside the irradiated field. Several randomized trials of concurrent chemoradiotherapy lend support to this concept (Table 30-1).

Adelstein et al.,⁹ Calais et al.,²² and Olmi et al.²³ reported randomized trials comparing conventional doses of RT with or without concurrent chemotherapy. Regardless of the specific chemotherapy regimens used, the trials demonstrated a significant and consistent benefit in local control rates, which translates into improvement in disease-free survival by a magnitude of 15% to 20%. The trials by Brizel et al.,²⁴ Budach et al.,²⁵ and Dobrowsky et al.²⁶ are noteworthy. The RT dose delivered in the chemoradiotherapy arms was intentionally lower than in the RT alone arms. Despite this apparent disadvantage, the chemoradiotherapy arms were still superior. These trials reinforce the concept that concurrent chemotherapy provides a synergistic enhancement of efficacy.

Single-agent Chemotherapy with Concurrent Radiotherapy

Platinum

Cisplatin is one of the most widely used radiosensitizers. The most impressive data on cisplatin as a radiosensitizer are from an NPC trial.²⁷ Significant improvement in 3-year overall survival was observed in the concurrent cisplatin with RT arm of 75%, compared to 44% in the conventional RT arm. Cisplatin is commonly delivered as a high-dose bolus. Alternate cisplatin schedules, including weekly or daily administration, have also been studied.

Definitive evidence on the efficacy of cisplatin in combination with RT was shown in a three-arm randomized trial, which compared conventional RT, concurrent cisplatin-RT, and concurrent cisplatin/5-FU with split-course RT.⁹ Cisplatin in the concurrent chemoradiotherapy arm was delivered at 100 mg per m² every 3 weeks. The chemotherapy split-course RT arm consisted of three cycles of cisplatin/5-FU every 4 weeks with concurrent chemoradiotherapy given only during the first and third cycles of chemotherapy. The 3-year overall survival were 23%, 37%, and 27% in the conventional RT, concurrent cisplatin-RT, and chemotherapy split-course RT arms, respectively. This trial showed that concurrent RT with high-dose bolus cisplatin significantly improved survival over RT alone. The loss of efficacy with split-course RT was not offset by the use of chemotherapy.

In order to improve tolerability of cisplatin, alternate scheduling of cisplatin has been investigated. Haselow et al.²⁸ randomized patients to conventional RT versus concurrent cisplatin (20 mg per m² weekly) with RT. No difference in outcome was observed, likely because the dose of cisplatin was too low to exert any significant sensitization or single-agent effect. A more recent retrospective review of patients with SCCHN treated with intensity-modulated radiotherapy and concurrent weekly cisplatin 30 mg/m²/week showed that it was a well-tolerated regimen and attained locoregional control rate of 86%, with overall survival 87% at 2 years.²⁹ Similarly, the University of Florida also reported its experience with hyperfractionated or concomitant boost RT delivered with concurrent cisplatin
(30 mg/m²/week).³⁰ The 5-year outcomes were encouraging with 83% of patients achieving locoregional control, 88% attaining distant metastasis-free survival, and almost 60% overall survival.

Table 30-1 Selected Randomized Concomittant Chemoradiotherapy Trials for LA-Nonmetastatic Head and Neck Cancer

Study	n	Regimen/Treatment Arm	Overall Survival	Disease-Free Survival
Single-agent chemoradiotherapy
Lo³⁴	136	A: RT	5-y A: 14%	2-y A: 18%
		B: 5-FU + RT	B: 32% P = NS	B: 49% P = .05
Browman³⁵	175	A: RT	2-y A: 50%	2-y A: 30%
		B: 5-FU + RT	B: 63% P =.08	B: 47% P = .06
Jeremic³¹	130	A: RT	5-y A: 46%	5-y A: 46%
		B: CDDP daily + HART	B: 25% P = .0075	B: 25% P = .0068
Dobrowsky²⁶	239	A: RT (70 Gy)	5-y A: 24%
		B: HART (55.3 Gy)	B: 31%
		C: MMC + HART (55.3 Gy)	C: 41% P < .05
Bonner⁴⁹	424	A: Cetuximab + RT	3-y A: 57%	NR
		B: RT	B: 44%
Multiagent chemoradiotherapy
Brizel²⁴	116	A: HART (75 Gy)	3-y A: 34%	3-y A: 41%
		B: CDDP, 5-FU + HART (70 Gy)	B: 55% P = .07	B: 61% P =.08
Wendt⁵¹	270	A: RT 3-y	A: 24% NR
		B: CDDP, 5-FU, LV + RT	B: 48% P < .0003
94-01 French Head and Neck Oncology and Radiotherapy Group^22,52	222	A: RT	5-y A: 16%	5-y A: 15%
		B: Carbo, 5-FU + RT	B: 22% P = .05	B: 27% P = .01
Olmi²³	192	A: RT 2-y A: 40%	2-y A: 23%
		B: Split-course accelerated fractionation RT	B: 37%	B: 20%
		C: Carbo, 5-FU + RT	C: 51% P = .13	C: 42% P = .02
Adelstein⁹	295	A: RT	3-y A: 23%	3-y A: 33%
		B: CDDP + RT	B: 37% P = .014	B: 51% P = .01
		C: CDDP, 5-FU + split-course RT	C: 27% P = NS	C: 41%
Garden⁵⁴ (phase II)	241	A: CDDP, 5-FU final 10 d + RT	2-y A: 57%	2-y A: 38%
		B: HU, 5-FU + RT	B: 69%	B: 49%
		C: PAC, CDDP + RT	C: 66%	C: 51%
Budach²⁵	384	A: 5-FU, MMC + HART (70.6 Gy)	5-y A: 29%	5-y A: 29%
		B: HART (77.6 Gy)	B: 24% P = .023	B: 27% P = .009
CDDP, cisplatin; PAC, paclitaxel; Carbo, carboplatin; MMC, mitomycin; RT, radiation therapy; HART, hyperfractionated radiation therapy; NS, not significant.

Daily platinum with concurrent RT was evaluated in a randomized study where patients received either RT or RT with cisplatin (6 mg/m²/day) or carboplatin (25 mg/m²/day).³¹ Patients who received concurrent chemoradiotherapy had significantly higher overall response rates, longer median survival (16 vs. 32 vs. 30 months), and 5-year survival rates (15% vs. 32% vs. 29%). There was no difference between the three treatment groups for distant control.

Huguenin et al. performed a randomized study of hyperfractionated RT alone or concurrent chemoradiotherapy where two cycles of cisplatin at 20 mg/m²/day for 5 days were given on weeks 1 and 5 of radiotherapy. Acute toxicity was similar in both arms. Locoregional control and distant disease-free survival were significantly improved with cisplatin.

Carboplatin has only been compared with cisplatin with concurrent RT in two studies.³²^,³³ Homma et al.³³ performed a phase II randomized study on 119 patients with LA-SCCHN, comparing RT with either weekly carboplatin (100 mg/m²/week) or daily cisplatin (4 mg/m²/day).³³ Although the 5-year overall survival rate did not significantly differ between treatments: 71% for carboplatin and 66% for cisplatin, the local control rate was 56% for the carboplatin arm and 36% for the cisplatin arm. Chitapanarux et al.³² performed a randomized noninferiority trial in 206 patients with LA-NPC comparing carboplatin (100 mg/m²/week) versus cisplatin (100 mg/m² every 3 weeks). Only 59% of patients in the cisplatin arm completed the planned concurrent chemoradiation treatment, compared with 73% in the carboplatin arm. The 3-year disease-free survival rates were 63% for the cisplatin group and 61% for the carboplatin group (P = .9613). The 3-year overall survival rates were 78% and 79% for cisplatin and carboplatin groups, respectively (P = .9884).

Antimetabolite

Bolus and infusional 5-FU can be combined with RT. Concurrent RT with bolus 5-FU (days 1 to 4 in the first week of RT then on Mondays, Wednesdays, and Fridays) compared to conventional RT significantly improved local control and survival in patients with unresectable SCCHN.³⁴ Continuous infusion of 5-FU at 1,200 mg/m²/day for 3 days on weeks 1 and 4 of RT resulted in higher complete response rate (68% vs. 56%) compared to RT alone.³⁵ A trend toward improved survival was seen with concurrent 5-FU and RT (median survival 33 vs. 25 months).

Taxanes

A variety of schedules for paclitaxel administration with concurrent RT have been evaluated. Twenty-four hour infusion of paclitaxel every 3 weeks during the course of RT showed high complete response rates to a maximum tolerated dose³⁶ of 75 mg per m². When paclitaxel (100 mg per m²) was given over 3 hours with RT every 3 weeks, a 90% response rate and median survival of 28 months were observed.³⁷ Paclitaxel (10.5 mg/m²/day) by continuous infusion for 7 weeks during RT is associated with high incidences of severe skin toxicity and mucositis.³⁸ Weekly paclitaxel over 1 hour with concurrent RT is feasible with tolerable toxicities.³⁹ Paclitaxel (120 mg per m²) administered as a 120-hour continuous infusion every 3 weeks during the course of RT resulted in a 70% complete response rate and median survival in excess of 50 months.⁴⁰ Paclitaxel can be safely combined with hyperfractionated RT with acceptable toxicities.⁴¹ For patients with poor performance status, concurrent paclitaxel-RT resulted in a response rate of 65% and a 2-year overall survival of 46%.⁴²

Docetaxel is commonly given weekly with RT.⁴³^,⁴⁴^,⁴⁵^,⁴⁶ The reported 3-year overall survival with weekly docetaxel (20 mg per m²) and RT in patients with LA-SCCHN is 47%.⁴³ In elderly patients, the radiosensitizing dose of docetaxel was found to be 12 mg/m²/week.⁴⁷

Biologic agents

Multiple novel targeted agents are currently being investigated in SCCHN. Epidermal growth factor receptor (EGFR) is commonly overexpressed in SCCHN. Two common modalities used to inhibit EGFR signaling are inhibition of tyrosine kinase ATP-binding sites by small molecule inhibitors, such as gefitinib and erlotinib, and inhibition of ligand-binding by monoclonal antibodies, such as cetuximab.

Cetuximab is administered with RT at a loading dose of 400 mg per m² followed by a weekly dose of 250 mg per m². This regimen was initially found to produce an overall response rate of 100%, including 87% complete responses.⁴⁸ The combination of RT and cetuximab versus RT alone was compared in a randomized trial in patients with LA-SCCHN.⁴⁹ The median duration of locoregional control was 24 months in the cetuximab-RT arm and 15 months in RT alone arm. The median survival was 49 months in the cetuximab-RT arm and 29 months in RT alone arm. Major toxicities were dermatitis, mucositis, dysphagia, and acneiform rash (in the cetuximab arm) and cetuximab did not exacerbate the common toxic effects associated with radiotherapy to the head and neck. Development of rash is associated with improved outcome.⁵⁰

Combination Chemotherapy with Concurrent Radiotherapy

Platinum and 5-FU combinations

One of the most common combination chemotherapy regimens to be delivered with concurrent RT is cisplatin and 5-FU (Table 30-1). Modulators of 5-FU such as leucovorin have been combined with cisplatin/5-FU and administered concurrently with RT.⁵¹ Wendt et al. evaluated patients with LA-SCCHN with RT alone or concurrent RT and cisplatin/5-FU/leucovorin. The 3-year overall survival rate was 24% in the RT arm and 48% in concurrent chemoradiotherapy arm.

Carboplatin and 5-FU delivered concurrently with RT was evaluated in two randomized studies. The French Head and Neck Oncology and Radiotherapy Group 94-01 study
compared hyperfractionated RT versus concurrent RT with carboplatin and 5-FU.²²^,⁵² Patients received a 4-day infusion of carboplatin (70 mg/m²/day) and 5-FU (600 mg/m²/day) for three cycles concurrently with RT. The 5-year overall survival in the hyperfractionated RT and concurrent chemoradiotherapy arms were 16% and 22%, respectively. Another three-arm randomized trial, compared arm A: conventional fractionation RT, arm B: split-course accelerated hyperfractionated RT, and arm C: conventional fractionation RT with concurrent carboplatin (75 mg/m²/day on days 1 to 4) and 5-FU (1,000 mg/m²/day over 96 hours on days 1 to 4) every 4 weeks.²³ The concurrent carboplatin-RT arm achieved the best disease-free survival among the three groups and trended toward improved overall survival.

Platinum-taxane combinations

Paclitaxel (up to doses of 40 mg/m²/week) has been successfully combined with cisplatin and RT after three cycles of paclitaxel-cisplatin induction chemotherapy.⁵³ This combination was one of the study arms in the RTOG 9703 trial, which compared three different concurrent chemoradiotherapy regimens (see below).⁵⁴

A phase II study, which evaluated induction carboplatin-paclitaxel followed by concurrent cisplatin (75 mg/m² every 3 weeks) and paclitaxel (30 mg/m²/week) with RT resulted in high hematologic toxicity rates.⁵⁵ The study was altered to utilize carboplatin (AUC1 weekly) and paclitaxel (30 mg/m²/week) with RT, which was better tolerated. The 2-year local control and overall survival rates were 82% and 71%, respectively. Despite the increase in toxicity with cisplatin-paclitaxel-RT, there were no significant differences in relapse-free survival or organ-preservation rates between concurrent regimens.

Weekly carboplatin and paclitaxel can be delivered with concurrent RT and achieves complete response rates between 60% and 75%.⁵⁶^,⁵⁷^,⁵⁸ Weekly carboplatin-paclitaxel can also be delivered concurrently with twice-daily hyperfractionated RT.⁵⁹

Cisplatin (20 mg per m²) and docetaxel (15 mg per m²) with concomitant boost RT resulted in a 2-year progression-free and overall survival rate of 61% and 65%.⁶⁰ Carboplatin and docetaxel have been successfully combined with concurrent RT in patients with unresectable SCCHN.⁶¹^,⁶² The overall response rate in one study was 97% with a complete response rate of 67%.⁶¹

Antimetabolite combinations

The FHX regimen consists of continuous infusional 5-FU on days 1 to 4 and hydroxyurea twice daily with daily RT on days 1 to 5 of a 2-week treatment cycle. A phase I/II trial evaluating the FHX regimen in poor-prognosis SCCHN patients showed significant antitumor activity.⁶³ The FHX regimen showed 5-year overall survival and locoregional control rates of 65% and 86%, respectively.⁶⁴

The RTOG 9703 trial, a three-arm randomized trial in patients with LA-SCCHN compared three different concurrent chemoradiotherapy regimens: arm A—cisplatin 10 mg/m2/day and 5-FU 400 mg/m2 continuous infusion daily for the final 10 days of RT, arm B—5-FU 800 mg/m2/day continuous infusion (F) and hydroxyurea 1,000 mg twice daily (H) every other week with RT (X) (FHX), and arm C—weekly paclitaxel 30 mg/m2 and cisplatin 20 mg/m2 with RT.⁵⁴ Overall survival rates among the three arms were 57% in arm A, 69% in arm B, and 67% in arm C. Toxicities and survival rates were not statistically different among the three arms. The patterns of failure were predominantly locoregional failure for all three arms.

In subsequent phase II trials, cisplatin and paclitaxel were each added to the FHX regimen and RT was delivered twice daily to further optimize locoregional control—CFHX,⁶⁵^,⁶⁶ T (1 hour) FHX⁶⁷, and T (120 hour) FHX.⁶⁸ All three trials showed 3-year survival rates of 60% or greater. The use of intensive chemoradiotherapy regimens (CFHX and TFHX) resulted in a reversal of the historical pattern of failure. Distant failure became the predominant site of treatment failure, even in the absence of surgery.⁶⁹

Combination Chemotherapy and Biologic Therapy with Concurrent Radiotherapy

When cetuximab was initially combined with cisplatin (100 mg per m² on weeks 1 and 4) and concurrent accelerated boost RT, the toxicity of the regimen was high despite encouraging results.⁷⁰ However, the encouraging efficacy result of this study prompted the development of RTOG 0522, which evaluated concurrent cisplatin-RT compared with concurrent cisplatin-cetuximab-RT in patients with unresectable LA-SCCHN. The study was completed in 2010 and results are pending. Cetuximab has also been combined with cisplatin (20 mg/m²/day) and fluorouracil (200 mg/m²/day) for 5 days delivered with three split courses of RT.⁷¹ Complete responses were observed in 70% of patients, but there was an unexpectedly high incidence of severe radiation dermatitis.

The University of Chicago performed a phase II trial, integrating gefitinib 250 mg daily into the FHX platform followed by 2 years of adjuvant gefitinib for LA-SCCHN.⁷² The complete response rate was 90%. The progression-free survival and overall survival at 4 years were 72% and 74%, respectively. High EGFR gene copy number was found to be associated with poorer outcome on this regimen. Gefitinib has also been combined with cisplatin (40 mg/m²/week) and concurrent RT.⁷³ Although this regimen was feasible, local control was only attained in 79% of patients. In contrast to the University of Chicago study, this study found that EGFR amplification tended to predict complete response to the treatment. Another study demonstrated the tolerability of paclitaxel (36 mg/m²/week) with gefitinib and RT; however, the response rates were not strikingly augmented in this small study.⁷⁴

The combination of cisplatin-erlotinib and RT can also be administered concurrently.⁷⁵ There was no dose-limiting toxicity when the dose of erlotinib was escalated to 150 mg daily.
The 3-year progression-free and overall survival rates were 61% and 72%, respectively.

Neoadjuvant Chemotherapy

The concept of neoadjuvant or induction chemotherapy was born when it was observed that treatment with one or two cycles of cisplatin⁷⁶ or methotrexate⁷⁷^,⁷⁸ in previously untreated patients with SCCHN consistently resulted in high response rates (50% to 70%). Combination chemotherapy further augmented the response rates to 80% with complete response rates as high as 50%.⁷⁹^,⁸⁰ Patients who achieve a complete response following neoadjuvant chemotherapy have a significant survival benefit over patients who only achieve a partial response or no response.⁸⁰^,⁸¹^,⁸² Among patients who achieve complete responses, those with microscopic residual disease after surgery have a poorer survival compared to patients with complete pathologic response.⁸³ Furthermore, the initial response to chemotherapy correlates with subsequent response to RT.⁸⁴ However, the response to chemotherapy is often transient and definitive therapy with surgery and/or RT is required.

The sensitivity of SCCHN to induction chemotherapy can be capitalized as an organ-preservation approach and/or to improve overall survival. Most neoadjuvant trials have focused on these two endpoints (organ-preservation rates and overall survival) (Table 30-2).

Jacobs et al.⁸⁵ initially demonstrated the feasibility of neoadjuvant chemotherapy by eliminating the need for radical surgery without compromising survival. Patients received three cycles of neoadjuvant cisplatin/5-FU, which is one of the most active chemotherapy regimens and has consistently showed high response rates (73% to 94%) across neoadjuvant studies.⁸¹^,⁸⁶^,⁸⁷^,⁸⁸ If a complete response was achieved, RT was administered as definitive therapy. However, those who did not achieve a complete response underwent surgical resection and adjuvant RT. Complete pathologic response was achieved in 40% of patients after induction therapy and these patients did not require surgery. The 2-year relapse-free and overall survivals were 60% and 70%, respectively, for patients who received neoadjuvant chemotherapy and definitive RT.

This lead to the Veterans Affairs (VA) Cooperative Study demonstrating equivalent survival rates between patients who received induction chemotherapy followed by RT and those who had laryngectomy.⁸⁹ This study randomized patients with LA-SCCHN to receive laryngectomy followed by RT or neoadjuvant chemotherapy and RT. Complete responders after two cycles of cisplatin/5-FU received a third cycle followed by RT; patients not responding after two cycles or with persistent disease following RT underwent salvage laryngectomy. Laryngeal preservation was feasible in 64% of patients in the neoadjuvant chemotherapy arm. An overall survival in both arms was 68% at 2-years. Fewer distant metastases but higher local recurrences were observed in the neoadjuvant chemotherapy group.

In line with the VA study, the EORTC also performed a study, which randomized patients with locally advanced pyriform sinus cancer to radical surgery followed by adjuvant RT or neoadjuvant cisplatin/5-FU for two cycles followed by RT.⁹⁰ Almost 60% of patients retained their larynx. Only 13% of patients who were randomized to the neoadjuvant chemotherapy arm required salvage laryngectomy. The 3-year survival was not significantly different between the two arms (neoadjuvant chemotherapy arm—57% and radical surgery arm—43%). Both trials confirmed that radical surgery can be substituted with chemotherapy-RT without compromising survival.

Developing in parallel with the neoadjuvant chemotherapy approach was concomitant chemoradiotherapy, which in itself is able to achieve high locoregional control rates. The two approaches were compared in the Intergroup 91-11.¹¹ The trial had three arms: (1) induction chemotherapy followed by RT, (2) concurrent chemoradiotherapy, and (3) daily singlefraction RT (Table 30-3). The concurrent chemoradiotherapy arm showed the best laryngectomy-free survival and local control rate. With the reporting of this trial, the neoadjuvant chemotherapy approach fell out of favor as an organ-preservation approach.

The survival benefit of neoadjuvant chemotherapy has not been consistently demonstrated, however, several meta-analyses have suggested that concurrent chemoradiotherapy confers a greater benefit compared to neoadjuvant chemotherapy in terms of overall survival.⁹¹^,⁹²^,⁹³^,⁹⁴ The recently updated MACH-NC meta-analysis reported a 0.96 hazard ratio (HR) of death in favor of neoadjuvant chemotherapy followed by locoregional therapy over locoregional therapy alone with an absolute benefit of 2.4% at 5 years.⁹⁴ However, when neoadjuvant chemotherapy was compared to concurrent chemoradiotherapy, the benefit favored concurrent chemoradiotherapy for overall survival (HR = 0.9) and event-free survival (HR = 0.8).⁹⁴

Until recently, the only studies that showed a survival benefit to neoadjuvant chemotherapy were the Groupe d’Etude des Tumeurs de la Tete et du Cou (GETTEC) study⁹⁵ and the Gruppo di Studio sui Tumori della Testa e del Collo (GSTTC) study.⁹⁶ Both studies used cisplatin/5-FU as the neoadjuvant chemotherapy regimen. The GETTEC trial showed that the median survival in the neoadjuvant chemotherapy arm was 5 years compared to 3 years in the RT alone arm.⁹⁵ The GSTTC trial initially found no difference between the two treatment groups with respect to overall survival (37% for the neoadjuvant arm vs. 29% for the local therapy arm at 2 years).⁹⁶ However, among patients with inoperable head and neck cancer, patients who received neoadjuvant chemotherapy had better 2-year overall survival (30% vs. 19%; p=.04) and locoregional control (84% vs. 72%; P = .05), and this benefit was maintained at 10 years.⁹⁷

Various efforts have been directed at improving the activity of neoadjuvant cisplatin/5-FU. The addition of a third chemotherapy agent to cisplatin/5-FU, such as methotrexate,⁹⁸ vinblastine,⁹⁹ or bleomycin,⁹⁹ did not increase its activity. Biologic modulators of 5-FU, such as leucovorin¹⁰⁰^,¹⁰¹^,¹⁰² and folinic acid,¹⁰³ have resulted in some improvement in complete response rates when combined with neoadjuvant cisplatin/5-FU in single-arm studies. Interferon added to cisplatin/5-FU/
leucovorin attained complete response rates around 60% with significant toxicity.¹⁰⁴^,¹⁰⁵^,¹⁰⁶

Table 30-2 Selected Neoadjuvant Chemoradiotherapy Trials for LA-Nonmetastatic Head and Neck Cancer

Study	Regimen/Treatment Arm	Primary Endpoint	Comments
Endpoint: laryngeal preservation
V.A. Laryngeal Cancer Study Group⁸⁹		Larynx preserv 2 y 64%	2 y surv
	A: RT		A: 68%
	B: CDDP, 5-FU × 3 → RT +/- Sx		B: 68% P = NS
Lefebvre⁹⁰		Larynx preserv 5 y 35%	3 y surv
	A: RT		A: 43%
	B: CDDP, 5-FU ×3 → RT +/- Sx		B: 57% P = NS
Richard²⁷¹		Larynx preserve 2 y 42%	2 y surv
	A: Sx → RT		A: 84%
	B: CDDP, 5-FU ×4 → RT		B: 69% P = .006
Forastiere¹¹		Larynx preserve 2 y	5 y surv
	A: RT	A: 70%	A: 56%
	B: CDDP, 5-FU × 4→ RT +/- Sx	B: 75% P = .005	B: 55%
	C: CDDP with RT	C: 88% P < .001	C: 54% P = NS
Endpoint: survival
Paccagnella GSTTC^96,97		10 y OS	Unresectable pts 10 y OS
	A: RT +/- Sx	A: 9%	A: 6%
	B: CDDP, 5-FU × 4 → RT +/- Sx	B:19% P = NS	B: 8% P = .04
Domenge GETTEC⁹⁵		Median survival	CR after chemotherapy: 20%
	A: RT+/- Sx	A: 3.3y
	B: CDDP, 5-FU × 3 → RT+/- Sx	B: 5.1y P = .03
Licitra²⁷²		OS 5-y	Mandible resection rate
	A: Sx	A: 55%	A: 52%
	B: CDDP, 5-FU × 3 → Sx RT for high risk disease	B: 55% P = NS	B: 31%
Vermorken EORTC 24971¹¹⁹		Median PFS	Median OS
	A: CDDP, 5-FU × 4 → RT	A: 8 mo	A: 15 mo
	B: DOC, CDDP, 5-FU × 4 → RT	B: 11 mo P = .007	B: 19 mo P = .02
Posner TAX324¹²⁰		Median PFS	Median OS
	A: CDDP, 5-FU × 4 → Carbo RT	A: 13 mo	A: 30 mo
	B: DOC, CDDP, 5-FU × 4 → Carbo RT	B: 36 mo P = .004	B: 71 mo P = .006
Sx, surgery; PFS, progression-free survival.

Both paclitaxel¹⁰⁷^,¹⁰⁸^,¹⁰⁹^,¹¹⁰ and docetaxel¹¹¹^,¹¹²^,¹¹³^,¹¹⁴^,¹¹⁵ have been utilized in neoadjuvant regimens. Hitt et al.¹⁰⁸^,¹⁰⁹ found that the combination of cisplatin 75 mg per m² and paclitaxel 300 mg per m² achieved an 80% overall response rate. Subsequent studies found that combining cisplatin/5-FU/paclitaxel further augmented the response rates to 88%, including 59% complete responses.¹¹⁶ This lead to a phase III trial comparing two neoadjuvant chemotherapy regimens: cisplatin/5-FU and
cisplatin/5-FU/paclitaxel.¹¹⁷ Although the overall response rate to neoadjuvant chemotherapy in the cisplatin/5-FU/paclitaxel arm was not significantly higher than the cisplatin/5-FU (88% vs. 78%), the complete response rate and the 2-year overall survival rate in the cisplatin/5-FU/paclitaxel arm were significantly improved (Table 30-3).

The combination of cisplatin/5-FU/docetaxel as a neoadjuvant regimen showed an overall response rate of 93%¹¹⁴ in one study but was poorly tolerated in another.¹¹⁸ Neoadjuvant cisplatin/5-FU/leucovorin/docetaxel (60 mg per m²) resulted in a 63% complete response rate and overall response rate of 93% following three cycles.¹¹² Higher doses of docetaxel (90 mg per m²) with cisplatin/5-FU/leucovorin did not improve response rates.¹¹³ The EORTC 24971 trial compared cisplatin/5-FU (PF) to cisplatin/5-FU/docetaxel (TPF) followed by RT. The triple induction regimen showed significantly improved progression-free and overall survival rates.¹¹⁹ Similarly, the TAX324 study randomized patients to receive neoadjuvant chemotherapy with TPF or PF followed by concurrent weekly low-dose carboplatin and RT.¹²⁰ The 3-year overall survivals were 62% in the TPF group and 48% in the PF group. There was better locoregional control in the TPF group, but the incidence of distant metastases did not differ between the two neoadjuvant regimens.

The efficacy of cisplatin and carboplatin may not be identical when combined with 5-FU in the neoadjuvant setting.¹²¹ In a randomized trial comparing neoadjuvant cisplatin/5-FU versus carboplatin/5-FU followed by definitive RT, the cisplatin/5-FU arm had higher response rates (92% vs. 76%) and better overall survival (5-year 49% vs. 25%). Although most neoadjuvant chemotherapy regimens used cisplatin/5-FU, induction chemotherapy with carboplatin-paclitaxel has also been utilized.¹¹⁰^,¹²² The overall response rate to induction carboplatin-paclitaxel approximated 90% with a 35% complete response rate.

Local and distant recurrences are commonly the cause of death for patients with LA-SCCHN. The distant disease control could potentially be enhanced by combining neoadjuvant chemotherapy to concurrent chemoradiotherapy. Paccagnella et al.¹²³ reported a phase II randomized study comparing definitive cisplatin/5-FU-RT versus neoadjuvant docetaxel/cisplatin/5-FU followed by cisplatin/5-FU-RT. The complete response rate after concurrent chemoradiotherapy was 21% in the definitive chemoradiotherapy arm versus 50% in the neoadjuvant therapy arm.

Only gold members can continue reading. Log In or Register to continue