Slow Accrual

Surgery with post operative radiotherapy (PORT) had been the mainstay of treatment for advanced laryngeal-pharyngeal cancers (ALHC) until the eighth decade of the past century. Total laryngectomy with post-operative radiotherapy (TL + PORT) used to result in permanent tracheostomy and loss of speech.

Early trials like the VA or European Organisation for Research and Treatment of Cancer (EORTC) trials compared surgery with post-operative radiotherapy to induction chemotherapy (ICT) and radiotherapy (RT). Subsequent attempts have been focused on the added benefit of including concurrent chemotherapy. There is no randomized trial available in the literature comparing concurrent chemoradiation with the standard treatment, i.e. surgery followed by radiotherapy. However, most of the studies comparing neoadjuvant chemotherapy and radiotherapy reported better locoregional control rates and better survival rates with surgery followed by post-operative chemotherapy. Further, the advances in primary voice rehabilitation have substantially improved the quality of life after laryngectomy. Thus, there is a strong case for comparing the results of concurrent chemo-radiation with surgery and post-operative radiotherapy in a randomized clinical trial. This trial will answer the question - "whether we are saving voice at the cost of life".

The investigators propose to randomize 900 patients of laryngeal and hypopharyngeal cancers in surgery with PORT and a concomitant chemoradiation arm and compare the survival and locoregional control rates.

TL + PORT has traditionally been the gold standard in management of ALHC. However, this results in permanent tracheostomy and a possible loss of speech. In case of partial laryngectomy and even in case of TL, there are various options of voice rehabilitation but the successes of all these procedures are highly variable. In 1980s several authors reported interesting possibility of LP with ICT. The first randomized study (RCT) came from VA group who randomized patients to receive either 2 cycles of ICT + RT Vs surgery +PORT. Patients with more than PR received a 3rd cycle followed by definitive RT. There were more local recurrences and fewer distant metastases in the ICT arm. Of the 166 ICT patients, nearly 1/3rd required salvage TL with ultimate LP in 66% surviving patients. These results proved that ICT and definitive RT can be effective in LP without compromising overall survival. European Organization for Research and Treatment of Cancer (EORTC) study randomly assigned hypopharynx cancer patients to receive either immediate surgery with PORT (arm 1) or ICT. Patients with a CR after 2 or 3 cycles of CT were treated there after by RT. Locoregional failures occurred at approximately the same frequencies in both arms but there were fewer distant failures in the ICT arm. The median survival was found to be similar in both arms with LPR of 35% in the ICT arm. This study showed the feasibility of LP in patients with cancer of the hypopharynx. A smaller trial from MSKCC15 reported 52% LPR and another European RCT reported poorer survival in the CT arm with LPR of 20% only. The latter trial had a smaller number of patients and imbalance in randomized groups (4 out of 5 stage IV patients got randomized into CT arm) that could have flawed the outcome. The 3 RCTs (excluding MSKCC trial) were compiled by MACH-NC to obtain a meta analysis that showed similar disease free survival and a non significant trend of higher 6% survival in pooled surgery arm which was counterbalanced by LPR of 58% in the pooled CT arm. Quality of life measures performed as part of the VA study demonstrated that LP offers better speech, good communication skills, lesser pain and depression compared to surgery.

In three arm study by RTOG (91-11), incidence of laryngectomy was 28% in induction (ICT) chemotherapy arm, 16% in concomitant arm, and 31% in radiation(RT) alone arm. Following TL, the incidence of major and minor complications ranged from 52% to 59% and did not differ significantly among the 3 arms. Fistula was lowest in RT alone arm (15%) and highest in concomitant arm (30%). Similar experience was reported from MSKCC with fistulas occurring in 39%, resulting in prolonged hospitalization. When compared with complication rates of surgery in untreated patients, the complication rates following unsuccessful LP protocol is significantly higher. In spite of higher morbidity, local-regional control is excellent for this group of patients. In RTOG trial, local-regional control following SS was 74% for CT arms and 90% for RT alone arm. At 24 months, the overall survival was equal in all arms.

The necessity of adding chemotherapy to radiotherapy itself is debatable. The MACH -NC reported 4% improvement in overall survival at 2 and 5 yrs with CT. So to prevent death of 400 patients at 5 years, 10,000 patients would have to undergo CT. In O'Sullivan' questionnaire based study, apart from extent of disease the other significant variables that influenced treatment recommendation were physicians specialty and their geographical area of practice. Most LP protocols are often accompanied by increased toxicity and are generally achieved in good performance status patients unlike majority of head and neck cancer patients. In VA trial, 77% patients had Karnofsky performance score (KPS) more than 80 and in RTOG 91-11 trial, 2/3 patients had KPS 90 or more. In RTOG trial, the mucosal toxicity in concurrent CT+RT arm was twice as much as the mucosal toxicity in other two arms. High grade toxic effects occurred more when CT was added to RT but there was no significant difference in rate of toxic effects between concurrent arm and ICT arm. Incidence of treatment modification, treatment interruption and hospitalizations are higher (compared to RT alone) when CT is administered concomitantly or during altered fractionation due to complications such as mucositis, dysphagia, pain, desquamation etc. The indirect costs attributable to non-surgical approaches e.g frequent expensive imagings, duration of treatment , duration of recuperation, cost of chemotherapy drugs, enhanced need for supportive care, stringent follow up and salvage surgery (in one third to half of the patients) may be more than the costs for radical surgery. Careful monitoring of the conservatively treated patient is mandatory to allow for early salvage of failures (in VA trial, induction CT arm had more local recurrences and only 2% patients were lost to follow-up). Given the infirmity and poor compliance of head and neck cancer patients such a stringent follow-up appears difficult (in VA trial patients were followed up every month for 1st year).

Nearly 40-60% patients fail on LPP and predicting this failure before spares these patients of unnecessary chemoradiation and its toxicities, trauma of recurrent disease and complications of salvage surgery. Mutation of the p53 gene has been found to regulate cell proliferation and chemosensitivity. LP is significantly higher in the group of patients whose tumors over expressed p53 but it does not predict survival. A retrospective study nested within the VA study reported that T stage, p53 over-expression and elevated proliferating cell nuclear antigen index were independent predictors of successful LP . Success of RT depends on killing all clonogenic cells that increases linearly with tumor volume (TV). Lesions are classified as T3 or T4 despite a wide variation in TV if one were to perform volumetric analysis for all. TV is one of the most precise and most relevant predictors of RT outcome. This inverse relationship may be explained on the basis of hypoxia due to central tumor necrosis that is detrimental for CT as well as RT. Cartilage invasion, soft tissue extension, volume, extensive nodal involvement, pre epiglottic space invasion, paraglottic space invasion and arytenoids infiltration are some of the radiological parameters that can predict poor outcome to RT.

Recent studies show adjuvant concurrent chemoradiation to the emerging standard of care for high risk tumors providing an estimated five year progression free survival benefit of 11% in advanced stage III and IV tumors or even early stage tumors with extranodal spread, positive resection margins, perineural involvement or vascular embolization. A similar study by RTOG showed an estimated 10% improvement in two year locoregional control in high risk tumors with multiple lymphnodal involvement, extranodal spread and positive resection margins.

To sum up, CT+RT has the advantages of potential radiosensitization by chemotherapy induced cell cycle redistribution, overcoming radio-resistance within the field of RT, targeting different subpopulation of cells leading to more kill, reduction or delay in distant metastases. Its disadvantages are increased expense, enhanced toxicity and need of good interdisciplinary integration. What needs to be appreciated is the fact that CT+RT has never been evaluated against the standard treatment of TL + PORT. Although the quality of life has been reported to be better after laryngeal preservation, speech rehabilitation has improved steadily over past decade. Time seems to be ripe now to compare LP with CT+RT with TL+PORT and speech rehabilitation with locoregional control and quality of life as endpoint.

DETAILED STUDY PLAN:

Study type: Prospective randomized controlled trial with 900 patients (450 in each arm). Trial size calculated for 392 events with expected improvement of base line survival of 42% by 10% (alpha error of 0.05 and power of 80).

RANDOMIZATION

Arm 1: Radiation Therapy+ CDDP

Arm 2: Surgery + Post operative RT(+ CDDP for high risk cases)

Arms 1 and 2: Cisplatin (CDDP) 100 mg/m2 over 20-30 minutes on days 1, 22, and 43. In arm 2 Cisplatin (CDDP)100 mg/m2 will be given to patients with multiple lymphnodal involvement, extranodal spread, positive resection margins, perineural involvement or vascular embolization.

Surgery: Near total/Total Laryngectomy with bilateral neck dissection with primary speech rehabilitation either by myo-mucosal shunt (NTL) or by primary tracheo-esophageal puncture.

Inclusion Criteria:

• Patients with biopsy-proven, previously untreated, T3/T4 tumors (with focal cartilage erosion on computed tomography [CT] scan); squamous cell carcinoma of larynx and hypopharynx.
• Patients with Karnofsky Performance Scale (KPS) > 80
• Patients must have resectable tumors which are potentially curable with conventional surgery and radiation therapy.
• Willing to participate in trial and get randomized

Exclusion Criteria:

• Gross cartilage invasion
• Extensive soft tissue infiltration
• Large nodal disease
• Distant metastases
• Synchronous primary