Study reference

Study characteristics

Patient characteristics

Intervention (I)

Comparison/control (C)

Follow-up

Outcome measures and effect size

Comments

1. Radiotherapy + cetuximab

Bonner (2006, 2010), Curran (2007)

Type of study:

Multicenter, randomized phase III trial

Setting and country:

73 centers in the US and 14 other countries (Spain, UK, South Africa, Poland, Belgium, France, Switzerland, Australia, Italy, Israel, New Zealand, Sweden, Germany, Netherlands)

Source of funding:

This study was funded by ImClone Systems, Bristol-Myers Squibb, and Merck KGaA.

Conflicts of interest:

Several authors report having been employed by and owning equity or stock options in the sponsors of this study (ImClone Systems). Another author is employed by a sponsor of this study (Merck). Several authors report having received consulting fees, having served on paid advisory boards, or having received lecture fees from ImClone Systems, Merck, or Bristol-Myers Squibb. One author has laboratory research support from ImClone Systems. The study was designed by ImClone Systems and the study chairman in collaboration with the lead investigators and was managed by ImClone Systems and Merck. ImClone Systems collected and analyzed the data. The article was written by the study chairman with assistance from the other authors.

Patients with locoregionally advanced squamous-cell carcinoma of the head and neck

Inclusion criteria:

• stage III or IV, non-metastatic, measurable squamous-cell carcinoma of the oropharynx, hypopharynx, or larynx
• medical suitability for definitive radiotherapy
• Karnofsky performance score ≥ 60
• normal hematopoietic, hepatic, and renal function

Exclusion criteria:

• had previously had cancer or had received chemotherapy within the preceding 3 y
• had undergone surgery or had previously received radiotherapy for head and neck cancer

N total at baseline:

Randomized: N = 424

I: N = 211

C: N = 213

Safety population: N = 420

I: N = 208

C: N = 212

Important characteristics:

Age, median (range)

I: 56 y (34-81)

C: 58 y (35-83)

Sex, n/N (%) male

I: 171/211 (81%)

C: 169/213 (79%)

Karnofsky performance score

60

I: 6/211 (3%)

C: 6/213 (3%)

70

I: 15/211 (7%)

C: 16/213 (8%)

80

I: 42/211 (20%)

C: 49/213 (23%)

90

I: 113/211 (54%)

C: 103/213 (49%)

100

I: 34/211 (16%)

C: 38/213 (18%)

Primary tumor site

Oropharynx

I: 118/211 (56%)

C: 135/213 (63%)

Larynx

I: 57/211 (27%)

C: 51/213 (24%)

Hypopharynx

I: 36/211 (17%)

C: 27/213 (13%)

Groups were comparable at baseline.

High-dose radiotherapy* + cetuximab (120-min infusion at an initial dose of 400 mg/m², followed by weekly 60-min infusions of 250 mg/m² for the duration of radiotherapy)

High-dose radiotherapy*

Length of follow-up:

Bonner (2006): median 54 months

Curran (2007): 12 months

Bonner (2010): median 60 months

Loss to follow-up or missing outcome data:

Bonner (2006): not reported

Curran (2007):

I: 4/211 (2%)

Reasons:

• QoL questionnaires not completed (n = 2)
• QoL questionnaires not assessable (n = 2)

C: 1/213 (< 1%)

Reason:

• QoL questionnaires not completed (n = 1)

Bonner (2010):

I: 4/211 (2%)

C: 4/213 (2%)

Reasons for loss to follow-up are not reported.

Based on subgroup analyses, cetuximab seemed to provide the most benefit for patients with oropharyngeal tumours.

Clinical outcomes

Overall survival

Months, median

I: 49.0 (95%CI: 32.8-69.5)

C: 29.3 (95%CI: 20.6-41.4)

HR 0.74 (95%CI: 0.57-0.97)

Overall survival at 3 years

I: 55%

C: 45%

Overall survival at 5 years

I: 46%

C: 36%

HR 0.73 (95%CI: 0.56-0.95)

Progression-free survival

Months, median (range)

I: 17.1

C: 12.4

HR 0.70 (95%CI: 0.54-0.90)

Progression-free survival at 2 years

I: 46%

C: 37%

Progression-free survival at 3 years

I: 42%

C: 31%

p = 0.04

Quality of life†

There were no significant differences in QoL scores between the treatment arms. This was particularly notable for global health status/QoL, social functioning, social eating, and social contact.

Safety

Acute toxic effects grade ≥ 3

Mucositis

I: 116/208 (56%)

C: 110/212 (52%)

p = 0.44

Acneiform rash

I: 35/208 (17%)

C: 3/212 (1%)

p < 0.001

Skin reaction

I: 73/208 (35%)

C: 45/212 (21%)

p < 0.05

Radiation dermatitis

I: 23%

C: 18%

p = 0.27

Weight loss

I I: 11%

C: 7%

p = 0.12

Xerostomia

I: 10/208 (5%)

C: 6/212 (3%)

p = 0.32

Dysphagia

I: 54/208 (26%)

C: 63/212 (30%)

p = 0.45

Asthenia

I: 4%

C: 5%

p = 0.64

Nausea

I: 2%

C: 2%

p = 1.00

Constipation

I: 5%

C: 5%

p = 1.00

Taste perversion

I: 0%

C: 0%

p = 1.00

Vomiting

I: 2%

C: 4%

p = 0.42

Pain

I: 6%

C: 7%

p = 0.84

Anorexia

I: 2%

C: 2%

p = 1.00

Fever

I: 1%

C: 1%

p = 1.00

Pharyngitis

I: 3%

C: 4%

p = 0.80

Dehydration

I: 6%

C: 8%

p = 0.57

Oral candidiasis

I: 0%

C: 0%

p = 1.00

Coughing

I: < 1%

C: 0%

p = 0.50

Voice alteration

I: 2%

C: 0%

p = 0.06

Diarrhea

I: 2%

C: 1%

p = 0.50

Headache

I: < 1%

C: < 1%

p = 1.00

Pruritus

I: 0%

C: 0%

p = 1.00

Infusion reaction

I: 6/208 (3%)

C: 0/212 (0%)

p = 0.01

Insomnia

I: 0%

C: 0%

p = 1.00

Dyspepsia

I: 0%

C: 1%

p = 0.50

Increased sputum

I: < 1 %

C: 1%

p = 0.62

Infection

I: 1%

C: 1%

p = 1.00

Anxiety

I: < 1%

C: 1%

p = 1.00

Chills

I: 0%

C: 0%

p = 1.00

Anemia

I: 1%

C: 6%

p = 0.006

Late toxic effects grade ≥ 3

Severe late effects related to radiation were reported in about 20% of the patients in each group. The sites most commonly affected were the esophagus, salivary glands, larynx, mucous membranes, subcutaneous tissues, bone, and skin.

Definitions:

* Radiotherapy regimens included:

• once daily: 70.0 Gy in 35 fractions; 2.0 Gy/fraction; 5 fractions/wk for 7 wk
• twice daily: 72.0-76.8 Gy in 60-64 fractions; 1.2 Gy/fraction; 10 fractions/wk for 6.0-6.5 wk
• concomitant boost: 72.0 Gy in 42 fractions; 32.4 Gy; 1.8 Gy/fraction; 5 fractions/wk for 3.6 wk; morning dose: 21.6 Gy; 1.8 Gy/fraction; 5 fractions/wk for 2.4 wk; afternoon dose: 18.0 Gy; 1.5 Gy/fraction; 5 fractions/wk for 2.4 wk

Uninvolved nodal areas of the neck were treated with 50 to 54 Gy, depending on the fractionation regimen used. Gross nodal disease received the same dose as the primary tumor. If, at registration, an investigator stipulated the need for neck dissection in patients with N2 or N3 disease of the neck (with such dissection formally recommended to take place 4 to 8 weeks after the completion of radiotherapy), the dose administered to the involved lymph nodes was 60 Gy. In the case of uncontrollable pain, a maximum of 2 five-day treatment breaks were allowed.

† The EORTC QLQ-C30 (version 3.0) and the QLQ-H&N35 instruments were used to assess QoL. For the QLQ-C30, 15 scales were derived from the initial 30 items: 5 functional scales (physical, role, emotional, cognitive, and social functioning), 3 symptom scales (fatigue, nausea and vomiting, pain), 6 single-item symptom scales (dyspnea, insomnia, appetite loss, constipation, diarrhea, financial difficulties), and 1 global health status/QoL scale. For the QLQ-H&N35, 18 scales were derived from the initial 35 items: 7 multi-item symptom scales (pain, swallowing, sensation, speech, eating from a social perspective, social interactions, and sexuality) and 11 single-item symptom scales (teeth, opening mouth, dry mouth, sticky saliva, coughing, felt ill, pain medication use, nutritional supplementation, feeding tube requirement, weight loss, and weight gain).

Remarks:

For patients with locoregionally advanced squamous-cell carcinoma of the head and neck, cetuximab plus radiotherapy significantly improves overall survival at 5 years compared with radiotherapy alone, confirming cetuximab plus radiotherapy as an important treatment option in this group of patients. Cetuximab-treated patients with prominent cetuximab-induced rash (grade 2 or above) have better survival than patients with no or grade 1 rash.

Authors conclusion:

Bonner (2006): Cetuximab plus radiotherapy is superior to radiotherapy alone in increasing both the duration of locoregional disease control and survival in locoregionally advanced head and neck cancer.

Curran (2007): The addition of cetuximab to radiotherapy significantly improved locoregional control and increased overall survival without adversely affecting quality of life.

Bonner (2010): Cetuximab plus radiotherapy is superior to radiotherapy alone in increasing both the duration of locoregional disease control and survival in locoregionally advanced head and neck cancer.

2. Radiotherapy + carboplatin

Fountzilas (2004)

Type of study:

Multicenter, randomized phase III trial

Setting and country:

2 hospitals in Greece, 1 hospital in Romania, 1 hospital in Germany

Source of funding:

Supported in part by Hellenic Cooperative Oncology Group research grant.

Conflicts of interest:

Not reported.

Patients with locally advanced non-nasopharyngeal head and neck cancer

Inclusion criteria:

• biopsy-proven previously untreated, stage III or IV (M0), squamous cell carcinoma of the oral cavity, oropharynx, hypopharynx, or larynx
• measurable or evaluable disease
• no synchronous primary tumors
• age ≥ 18 y
• ECOG performance status 0-2
• adequate bone marrow, hepatic, and renal function
• CT or MRI scan of the head and neck region within 2-3 wk prior to initiation of treatment
• adequate cardiovascular, pulmonary, nutritional and mental status

Exclusion criteria:

Not reported.

N total at baseline:

Randomized: N = 79

I: N = 38

C: N = 41

Important characteristics:

Age, median (range)

I: 53 y (31-71)

C: 58 y (40-78)

Sex, n/N (%) male

I: 29/38 (76%)

C: 36/41 (88%)

Performance status

0

I: 24/38 (63%)

C: 22/41 (54%)

1

I: 13/38 (34%)

C: 15/41 (37%)

2

I: 1/38 (3%)

C: 2/41 (5%)

Primary tumor site

Larynx

I: 15/38 (39%)

C: 18/41 (44%)

Oropharynx

I: 12/38 (32%)

C: 14/41 (34%)

Hypopharynx

I: 4/38 (10%)

C: 5/41 (12%)

Oral cavity

I: 7/38 (18%)

C: 4/41 (10%)

There were no significant differences in major patient or tumor characteristics between the treatment groups.

Standard fractionated radiotherapy* (total dose of 70 Gy at 1.8 Gy/fraction per day, 5 days/wk) + carboplatin (AUC of 7 on days 2, 22 and 42)

Standard fractionated radiotherapy* (total dose of 70 Gy at 1.8 Gy/fraction per day, 5 days/wk)

Length of follow-up:

Median 5 years (range: 0.02-7.2+)

Loss to follow-up or missing outcome data:

I: 2/38 (5%)

C: 1/41 (2%)

Reasons for loss to follow-up are not reported.

Clinical outcomes

Overall survival

Months, median (range)

I: 24.5 (0.2-79.9)

C: 12.2 (1.2-81.7)

Median overall survival curves were compared using the log-rank test: p = 0.0064

Overall survival at 3 years

I: 42%

C: 17.5%

Overall survival at 5 years

I: 38%

C: 9%

Overall survival curves were compared using the log-rank test: p < 0.001

Safety

Adverse events grade ≥ 3

Anemia

I: 3/38 (8%)

C: 0/40 (0%)

p ≥ 0.05

Leukopenia

I: 7/38 (18%)

C: 1/40 (3%)

p ≥ 0.05

Neutropenia

I: 1/38 (3%)

C: 1/40 (3%)

p ≥ 0.05

Thrombocytopenia

I: 10/38 (26%)

C: 0/40 (0%)

p = 0.0004

Nausea/vomiting

I: 6/38 (16%)

C: 0/40 (0%)

p = 0.0107

Stomatitis

I: 7/38 (18%)

C: 13/40 (33%)

p ≥ 0.05

Diarrhea

I: 0/40 (0%)

C: 1/40 (3%)

p ≥ 0.05

Mouth dryness

I: 1/38 (3%)

C: 0/40 (0%)

p ≥ 0.05

Dermatitis

I: 0/40 (0%)

C: 1/40 (3%)

p ≥ 0.05

Hoarseness

I: 1/38 (3%)

C: 1/40 (3%)

p ≥ 0.05

Dysphagia

I: 3/38 (8%)

C: 2/40 (5%)

p ≥ 0.05

Neurotoxicity

I: 1/38 (3%)

C: 0/40 (0%)

p ≥ 0.05

Weight loss

I: 2/38 (5%)

C: 3/40 (8%)

p ≥ 0.05

Infection

I: 1/38 (3%)

C: 0/40 (0%)

p ≥ 0.05

Definitions:

* RT was delivered using linear accelerator or cobalt-60 unit with a source-to-surface or source-to-isocenter distance ≥ 80 cm. A total dose of 70 Gy was given using standard fractionatron at 1.8 Gy/fraction/d, 5 d/week. In lateral opposing fields, dose was specified at midplane, while in the anterior field, dose was calculated at 3 cm depth. The clinically uninvolved areas received 50 Gy, while enlarged lymph nodes were irradiated with an additional boost of 20 Gy. The primary treatment fields were reduced off the spinal cord at 50 Gy.

Remarks:

-

Authors conclusion:

The present study clearly demonstrated that concomitant conventional radiotherapy and carboplatin, given in a 3-wk schedule, significantly improved 3-yr survival and locoregional control as compared to that achieved by conventional radiotherapy alone. This survival benefit was accompanied by considerable incidence of severe toxicity, mainly myelotoxicity and nausea/vomiting.

Jeremic (1997)

Type of study:

Single-center, randomized trial

Setting and country:

Yugoslavia

Source of funding:

Not reported.

Conflicts of interest:

Not reported.

Patients with locally advanced, unresectable head and neck cancer

Inclusion criteria:

• histologically confirmed locally advanced unresectable non-metastatic (stage III and IV) squamous cell carcinoma of the head and neck
• Karnofsky performance score ≥ 50
• age > 18 y
• adequate hematological, renal and hepatic function,
• no previous therapy
• measurable tumor mass
• no serious concomitant diseases

Exclusion criteria:

Not reported.

N total at baseline:

Randomized: N = 106

I: N = 53

C: N = 53

Important characteristics:

Age, median (range)

I: 58 y (34-68)

C: 59 y (35-70)

Sex, n/N (%) male

I: 45/53 (85%)

C: 43/53 (81%)

Performance status

80-100

I: 40/53 (75%)

C: 37/53 (70%)

50-70

I: 13/53 (25%)

C: 16/53 (25%)

Primary tumor site

Oral cavity

I: 8/53 (15%)

C: 9/53 (17%)

Oropharynx

I: 20/53 (38%)

C: 21/53 (40%)

Larynx

I: 9/53 (17%)

C: 8/53 (15%)

Hypopharynx

I: 10/53 (19%)

C: 10/53 (19%)

Nasopharynx

I: 6/53 (11%)

C: 5/53 (9%)

Groups were comparable at baseline.

Standard fractionated radiotherapy* (total dose of 70 Gy at 1.8-2.0 Gy/fraction per day, 5 days/wk) + carboplatin (bolus of 25 mg/m² infused daily 45-60 min before irradition)

Standard fractionated radiotherapy* (total dose of 70 Gy at 1.8-2.0 Gy/fraction per day, 5 days/wk)

Length of follow-up:

Not reported.

Loss to follow-up or missing outcome data:

3 patients were lost to follow-up immediately after the treatment completion, but were included in the final analysis, excluding that of the late treatment-related toxicity. However, it is unclear to what treatment group these patients were allocated.

Clinical outcomes

Overall survival

Months, median

I: 30

C: 16

Overall survival at 1 year

I: 76%

C: 57%

Overall survival at 2 years

I: 55%

C: 35%

Overall survival at 3 years

I: 47%

C: 27%

Overall survival at 4 years

I: 31%

C: 17%

Overall survival at 5 years

I: 29%

C: 15%

Overall survival curves were compared using the log-rank test: p = 0.019

Safety

Acute toxic effects grade ≥ 3

Stomatitis

I: 7/53 (13%)

C: 5/53 (9%)

p ≥ 0.05

Xerostomia

I: 1/53 (2%)

C: 3/53 (6%)

p ≥ 0.05

Esophagitis

I: 2/53 (4%)

C: 2/53 (4%)

p ≥ 0.05

Nausea/vomiting

I: 1/53 (2%)

C: 0/53 (0%)

p ≥ 0.05

Nephrotoxicity

I: 0/53 (0%)

C: 0/53 (0%)

p = 1.00

Leukopenia

I: 6/53 (11%)

C: 0/53 (0%)

p = 0.012

Thrombocytopenia

I: 4/53 (8%)

C: 0/53 (0%)

p = 0.041

Late toxic effects grade ≥ 3

Subcutaneous tissue

I: 1/51 (2%)

C: 1/51 (2%)

p = 1.00

Bone

I: 1/51 (2%)

C: 1/51 (2%)

p = 1.00

Skin

I: 1/51 (2%)

C: 1/51 (2%)

p = 1.00

Definitions:

* RT was employed with 6-10 MV photons from linear accelerators. The target volume included the primary tumor, the lymph nodes of the neck and supraclavicular fossa. The tumor bearing area received 70 Gy and uninvolved neck and supraclavicular nodes received 45 Gy. Daily fractions of 1.8-2.0 Gy were used five times a week. The primary tumor and upper neck nodes were treated with two lateral opposed fields with 45 Gy in 5 weeks after which reduced lateral fields were used to boost the primary and the involved nodes to 70 Gy in 7-7.5 weeks. The dose to the spinal cord was kept at 45 Gy. The uninvolved lower neck and supraclavicular nodes were treated with a single anterior field, starting 0.5 cm below the lateral fields and with a total dose of 45 Gy at 3-3.5 cm depth in 5 weeks with apical lung shield. All fields were treated every day.

Remarks:

Initially, 85 patients were thought to be required per arm to detect a difference in the 3-year survival rate of 20% with a significance level of P < 0.05 and a power of 0.8 assuming a baseline survival rate of 25%. However, the study ended in December 1990 before the patient accrual had reached this number, because the chief investigator had to leave the department at this time.

Authors conclusion:

This rather small, single-institutional study showed promising results and acceptable toxicity in patients with heterogenous head and neck subsites treated with standard fraction radiotherapy and concurrent low-dose daily carboplatin over that obtained with the same radiotherapy given alone. Improvement in survival was due to the improvement in local tumor control that was significantly superior for the intervention group attributed to the effect of carboplatin when given daily during radiotherapy, as in this schedule.

Ruo Redda (2010)

Type of study:

Multicenter, randomized phase III trial

Setting and country:

6 centers in Italy

Source of funding:

Not reported.

Conflicts of interest:

Not reported.

Patients with locally advanced, unresectable head and neck cancer

Inclusion criteria:

• biopsy-proven diagnosis of locally advanced and unresectable stage III or IV non-metastatic head and neck squamous cell carcinoma
• tumor/lymph node mass had to be measurable by diagnostic imaging
• no prior chemotherapy or radiotherapy for any kind of cancer (except for non-melanoma skin cancer or in situ cervical cancer)
• age 18-70 y
• ECOG performance status ≤ 2
• no serious concomitant diseases
• adequate bone marrow reserve, renal function and liver function
• adequate nutritional and liquid intake

Exclusion criteria:

Not reported.

N total at baseline:

Randomized: N = 164

I: N = 82

C: N = 82

Important characteristics:

Age, median (range)

I: 58 y (39-70)

C: 61 y (40-70)

Sex, n/N (%) male

I: 66/73 (90%)

C: 63/72 (88%)

Performance status

0

I: 47/73 (64%)

C: 30/72 (42%)

1

I: 18/73 (25%)

C: 33/72 (46%)

2

I: 8/73 (11%)

C: 9/72 (13%)

Primary tumor site

Oral cavity

I: 14/73 (19%)

C: 14/72 (19%)

Oropharynx

I: 42/73 (58%)

C: 39/72 (54%)

Larynx

I: 9/73 (12%)

C: 12/72 (14%)

Hypopharynx

I: 8/73 (11%)

C: 7/72 (10%)

Groups were comparable at baseline, except for ECOG performance status. The percentage of patients with an ECOG performance status of 1 was higher in the control group, whereas the percentage of patients with an ECOG performance status of 0 was higher in the intervention group.

Standard fractionated radiotherapy* (total dose of 70 Gy at 2 Gy/fraction, 5 days/wk) + carboplatin (bolus of 45 mg/m² infused 45-60 min before irraditation, on days 1-5 of weeks 1, 3, 5 and 7; total dose of 900 mg/m²)

Standard fractionated radiotherapy* (total dose of 70 Gy at 2 Gy/fraction, 5 days/wk)

Length of follow-up:

Median 26.2 months (range: 6.2-169.5)

Loss to follow-up or missing outcome data:

None.

Clinical outcomes

Overall survival at 3 years

I: 28.9%

C: 11.1%

Overall survival at 5 years

I: 9%

C: 6.9%

Overall survival at 10 years

I: 5.5%

C: 6.9%

Overall survival curves were compared using the log-rank test: p = 0.02

Disease-free survival at 3 years

I: 16%

C: 9%

Disease-free survival at 5 years

I: 6.8%

C: 5.5%

Disease-free survival at 10 years

I: 6.8%

C: 5.5%

Disease-free survival curves were compared using the log-rank test: p = 0.09

Safety

Acute toxic effects grade ≥ 3

Anemia

I: 3/73 (4%)

C: 0/72 (0%)

p ≥ 0.05

Leukopenia

I: 7/73 (10%)

C: 0/72 (0%)

p ≥ 0.05

Thrombocytopenia

I: 0/73 (0%)

C: 0/72 (0%)

p = 1.00

Mucositis

I: 10/73 (14%)

C: 9/72 (13%)

p ≥ 0.05

Late toxic effects grade ≥ 3

The prevalence of late toxic effects was similar between treatment groups, except for severe neck fibrosis, which occurred more frequently in patients treated with radiotherapy and carboplatin (7 versus 3 cases). One patient treated with radiotherapy and carboplatin developed necrosis of the mandibular bone. No radiation myelitis or toxic-related death was observed in either treatment group.

Definitions:

* Radiotherapy was delivered by the use of cobalt-60 gamma rays or 6 MV photons and with electron beams to give a boost to the posterior cervical lymph nodes.

The planned total radiation dose was 70 Gy to the involved areas given with a conventional fractionated schedule of 2 Gy/die, 5 fractions/week. Each patient was simulated before treatment, and field sizes were individualized using custom blocking to spare normal tissues. The primary tumor and upper neck nodes were treated with two opposed lateral fields; at the dose of 42 Gy, the lateral fields were reduced for spinal cord protection. Dose variation across the tumor target volume in the central plane was limited to 10% of the prescribed dose, from minimum to maximum, whenever possible. The uninvolved lower neck and supraclavicular nodal regions were treated with a single anterior field with a total dose of 54 Gy, using apical lung shields. A central block was used, when required.

Remarks:

-

Authors conclusion:

Long-term results in both treatment arms of the trial appear less positive than most published series. However, our findings do not exclude that carboplatin

may be beneficial, but the benefit in local control must be lower than the 15% assumed to dimension the trial.

3. Radiotherapy + carboplatin and 5-FU

Bourhis (2012)

(GORTEC 99-02 trial)

Type of study:

Multicenter, open-label, randomized phase III trial

Setting and country:

22 university hospitals, cancer centers, and private hospitals in France and Belgium

Source of funding:

French Ministry of Health. The sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

Conflicts of interest:

The authors declare that they have no conflicts of interest.

Patients with locally advanced head and neck squamous-cell carcinoma

Inclusion criteria:

• previously untreated histologically proven squamous-cell carcinoma of the oral cavity, oropharynx, hypopharynx or larynx
• stage III or IV disease (excluding distant metastases)
• ECOG performance status 0-2
• adequate hematological, renal, and liver function
• no previous radiotherapy or chemotherapy
• no history of other cancer in the previous 5 years
• no clinically significant cardiac disease
• no clinically significant impairment that would have contraindicated the use of chemotherapy

Exclusion criteria:

Not reported.

N total at baseline:

Randomized: N = 840

I1: N = 279

I2: N = 280

C: N = 281

Important characteristics:

Age, mean (SD, range)

I1: 56.1 y (7.7, 39-77)

I2: 56.9 (8.2, 37-75)

C: 56.5 y (8.6, 34-74)

Sex, n/N (%) male

I1: 244/279 (87%)

I2: 245/280 (88%)

C: 242/281 (86%)

Primary tumor site

No primary

I1: 244/279 (87%)

I2: 245/280 (88%)

C: 242/281 (86%)

Oropharynx

I1: 244/279 (87%)

I2: 245/280 (88%)

C: 242/281 (86%)

Oral cavity

I1: 244/279 (87%)

I2: 245/280 (88%)

C: 242/281 (86%)

Hypopharynx

I1: 244/279 (87%)

I2: 245/280 (88%)

C: 242/281 (86%)

Larynx

I1: 244/279 (87%)

I2: 245/280 (88%)

C: 242/281 (86%)

Groups were comparable at baseline, except for ECOG performance status. The percentage of patients with an ECOG performance status of 1 was higher in the control group, whereas the percentage of patients with an ECOG performance status of 0 was higher in the intervention group.

Intervention 1:

Standard fractionated radiotherapy* (total dose of 70 Gy in 7 wk at 5 fractions of 2 Gy per wk, with spinal cord exclusion at 40 Gy) + carboplatin (70 mg/m² per day on day 1-4, day 22-25 and day 43-46) and fluorouracil (600 mg/m² per day on day 1-4, day 22-25 and day 43-46)

Intervention 2:

Accelerated radiotherapy* (total dose of 70 Gy in 6 wk at 5 fractions of 2 Gy per wk until 40 Gy, with spinal cord exclusion at 40 Gy, followed by 1.5 Gy per fraction twice daily for 5 days/wk for the remaining 30 Gy) + carboplatin (70 mg/m² per day on day 1-5 and day 29-33) and fluorouracil (600 mg/m² per day on day 1-5 and day 29-33)

Very accelerated radiotherapy* (total dose of 64.8 Gy in 3.5 wk at 1.8 Gy per fraction twice daily for 5 days/wk, with spinal cord exclusion at 34.2 Gy)

Length of follow-up:

Median 5.2 years (range: 0.4-9.5; IQR: 4.9-6.2)

Loss to follow-up or missing outcome data:

I1: 1 patient had ≤ 3 years follow-up, 278 patients had > 3 years follow-up

I2: 4 patients had ≤ 3 years follow-up, 276 patients had > 3 years follow-up

C: 1 patient had ≤ 3 years follow-up, 280 patients had > 3 years follow-up

It is not reported why 6 patients were followed up for less than 3 years.

Clinical outcomes

Overall survival at 3 years

I1: 42.6% (95%CI: 37.0-48.5)

I2: 39.4% (95%CI: 33.8-45.3)

C: 36.5% (95%CI: 31.1-42.3)

I1 versus C: HR 0.81 (95%CI: 0.67-0.99)

I2 versus C: HR 0.87 (95%CI: 0.72-1.06)

I2 versus I1: HR 1.05 (95%CI: 0.86-1.29)

Progression-free survival at 3 years†

I1: 37.6% (95%CI: 32.1-43.4)

I2: 34.1% (95%CI: 28.7-39.8)

C: 32.2% (95%CI: 27.0-37.9)

I1 versus C: HR 0.82 (95%CI: 0.67-0.99)

I2 versus C: HR 0.83 (95%CI: 0.69-1.01)

I2 versus I1: HR 1.02 (95%CI: 0.84-1.23)

Safety

Acute toxic effects grade ≥ 3 (during treatment)

At least one acute toxic effect

I1: 222/266 (83%)

I2: 243/276 (88%)

C: 243/274 (89%)

p = 0.07

Mucositis (RTOG grading)

I1: 180/262 (69%)

I2: 205/271 (76%)

C: 226/268 (84%)

p = 0.0001

Mucositis (WHO grading)

I1: 206/264 (78%)

I2: 228/271 (84%)

C: 237/267 (89%)

p = 0.016

Skin toxicity (RTOG grading)

I1: 108/260 (42%)

I2: 118/268 (44%)

C: 104/265 (37%)

p = 0.24

Feeding tube

I1: 158/265 (60%)

I2: 176/276 (64%)

C: 190/272 (70%)

I1 versus C: p = 0.013

I2 versus C: p = 0.13

I2 versus I1: p = 0.32

All hematological toxicity (during chemotherapy)

I1: 36/264 (14%)

I2: 45/275 (16%)

p = 0.57

Anemia (during chemotherapy)

I1: 11/264 (4%)

I2: 7/275 (3%)

p = 0.48

Leukopenia (during chemotherapy)

I1: 28/264 (11%)

I2: 38/275 (14%)

p = 0.55

Thrombocytopenia (during chemotherapy)

I1: 10/264 (4%)

I2: 13/275 (5%)

p = 0.90

Late toxic effects grade ≥ 3 (during follow-up; 1-5 years)

Feeding tube

I1: 36% at 1 y, 16% at 2 y, 11% at 3 y, 8% at 4 y, and 13% at 5 y

I2: 39% at 1 y, 17% at 2 y, 11% at 3 y, 7% at 4 y, and 6% at 5 y

C: 43% at 1 y, 23% at 2 y, 18% at 3 y, 14% at 4 y, and 25% at 5 y

I1 versus C: p = 0.027

I2 versus C: p = 0.092

I2 versus I1: p = 0.59

Xerostomia

I1: 8% at 1 y, 10% at 2 y, 7% at 3 y, 8% at 4 y, and 4% at 5 y

I2: 11% at 1 y, 6% at 2 y, 4% at 3 y, 3% at 4 y, and 7% at 5 y

C: 9% at 1 y, 8% at 2 y, 8% at 3 y, 5% at 4 y, and 6% at 5 y

p ≥ 0.05

Subcutaneous fibrosis

I1: 5% at 1 y, 8% at 2 y, 13% at 3 y, 7% at 4 y, and 5% at 5 y

I2: 5% at 1 y, 5% at 2 y, 3% at 3 y, 9% at 4 y, and 9% at 5 y

C: 7% at 1 y, 7% at 2 y, 5% at 3 y, 7% at 4 y, and 6% at 5 y

p ≥ 0.05

Mucosal toxicity

I1: 5% at 1 y, 8% at 2 y, 5% at 3 y, 3% at 4 y, and 1% at 5 y

I2: 5% at 1 y, 2% at 2 y, 0% at 3 y, 1% at 4 y, and 0% at 5 y

C: 5% at 1 y, 3% at 2 y, 1% at 3 y, 1% at 4 y, and 2% at 5 y

p ≥ 0.05

Laryngeal toxicity

I1: 6% at 1 y, 5% at 2 y, 4% at 3 y, 3% at 4 y, and 3% at 5 y

I2: 3% at 1 y, 2% at 2 y, 1% at 3 y, 0% at 4 y, and 1% at 5 y

C: 5% at 1 y, 3% at 2 y, 3% at 3 y, 2% at 4 y, and 0% at 5 y

p ≥ 0.05

Bone toxicity

Rate of late bone toxicity did not differ between treatment groups (data not shown).

Definitions:

* We used a 4-6 MV linac and a conventional three-dimensional treatment planning system (no intensity modulated radiation therapy was done at the same time). When appropriate, cervical posterior nodes were treated with electron beams (8-12 MeV) or with oblique posterior photon beams. The prophylactic nodal irradiation dose was 45 Gy in the uninvolved neck in the very accelerated radiotherapy group and 50 Gy for the two other groups.

† Progression-free survival was defined as time between randomisation and the first of the following events: locoregional progression or relapse, distant relapse, or death from any cause (or the last follow-up contact for patients who did not have any of these events).

d

Remarks:

-

Authors conclusion:

Chemotherapy has a substantial treatment effect given concomitantly with radiotherapy and acceleration of radiotherapy cannot compensate for the absence of chemotherapy. We noted the most favourable outcomes for conventional chemoradiotherapy, suggesting that acceleration of radiotherapy is probably not beneficial in concomitant chemoradiotherapy schedules.

Calais (1999), Denis (2004)

(GORTEC 94-01 trial)

Type of study:

Multicenter, randomized phase III trial

Setting and country:

university hospitals, cancer centers, and private hospitals in France

Source of funding:

French Ministry of Health.

Conflicts of interest:

The authors indicated no potential conflicts of interest.

Patients with advanced oropharynx carcinoma

Inclusion criteria:

• invasive squamous cell carcinoma of the oropharynx (stage III or IV, without evidence of distant metastases)
• age < 75 y
• Karnofsky performance score ≥ 60
• neutrophil count > 1500 cells/mm³
• platelet count > 120.000 cells/mm³
• serum creatinine concentration ≤ 1.4 mg/dL (120 mmol/L)

Exclusion criteria:

• body weight loss > 20%
• previous treatment for this disease or any other cancer (except basal cell carcinoma of the skin)
• synchronous primary lesions

N total at baseline:

Randomized: N = 222

I: N = 113

C: N = 109

Important characteristics:

Age, mean (range)

I: 55.7 y (32-73)

C: 54.4 y (34-74)

Sex, n/N (%) male

I: 99/109 (91%)

C: 101/113 (89%)

Karnofsky performance score

70

I: 15/109 (14%)

C: 16/113 (14%)

80

I: 36/109 (33%)

C: 25/113 (22%)

90-100

I: 58/109 (53%)

C: 72/113 (64%)

Primary tumor site

Tonsillar region

I: 43/109 (39%)

C: 42/113 (37%)

Base of tongue

I: 40/109 (37%)

C: 39/113 (34%)

Soft palate-uvula

I: 12/109 (11%)

C: 11/113 (10%)

Posterior wall

I: 8/109 (7%)

C: 7/113 (6%)

Non-classified

I: 5/109 (5%)

C: 4/113 (4%)

Groups were comparable at baseline, except for a slightly higher proportion of patients with N3 lymph nodes in the intervention. Also, the intervention group comprised more patients with a Karnofsky performance score of 80 and less patients with a Karnofsky performance score of 90-100 than the control group.

Standard fractionated radiotherapy* (total dose of 70 Gy in 35 fractions at 2 Gy/fraction per day, 5 days/wk) + carboplatin (i.v. bolus of 70 mg/m² per day for 4 days) and fluorouracil (600 mg/m² per day infused continously in 24 hr for 4 days)

Standard fractionated radiotherapy* (total dose of 70 Gy in 35 fractions at 2 Gy/fraction per day, 5 days/wk)

Length of follow-up:

Calais (1999): median 35 months (range: 12-56)

Denis (2004): median 5.5 years (range: 4-7.2)

Loss to follow-up or missing outcome data:

No patient was lost to follow-up for assessment of survival. Three were lost to follow-up for the late toxicity assessment, but were free of local or distant recurrence (2 patients in intervention group and 1 patient in the control group).

Clinical outcomes

Overall survival

Months, median

Calais (1999):

I: 29.2

C: 15.4

Denis (2004):

I: 20

C: 13

Overall survival at 3 years

I: 51% (95%CI: 39-68)

C: 31% (95%CI: 18-49)

Overall survival curves were compared using the log-rank test: p = 0.02

Disease-free survival at 3 years

I: 42% (95%CI: 30-57)

C: 20% (95%CI: 10-33)

Disease-free survival curves were compared using the log-rank test: p = 0.04

Overall survival at 5 years

I: 22.4%

C: 15.8%

Overall survival curves were compared using the log-rank test: p = 0.05

Disease-free survival at 5 years

I: 26.6%

C: 14.6%

Disease-free survival curves were compared using the log-rank test: p = 0.01

Local recurrence

Calais (1999):

I: 36/109 (33%)

C: 58/113 (51%)

p = 0.0067

Denis (2004):

I: 45/109 (41%)

C: 66/113 (58%)

p = 0.0155

Safety

Acute toxic effects grade ≥ 3

Mucositis

I: 71% (95%CI: 54-85)

C: 39% (95%CI: 29-56)

p = 0.005

Radiation dermatitis

I: 67%

C: 59%

p = 0.02

Weight loss

I: 50%

C: 21%

p < 0.05

Hematology

Neutropenia

I: 4%

C: 0%

p = 0.04

Thrombocytopenia

I: 6%

C: 1%

p = 0.04

Anemia

I: 3%

C: 0%

p = 0.05

Toxic death

I: 1/109 (1%)

C: 0/113 (0%)

Late toxic effects grade ≥ 3

Calais (1999):

Xerostomia

I: 10/109 (9%)

C: 6/113 (5g%)

p = 0.1

Cervical fibrosis

I: 12%

C: 3%

p = 0.08

Bone necrosis

I: 0%

C: 0%

p = 1.00

Radiation myelitis

I: 0%

C: 0%

p = 1.00

Denis (2004):

≥ 1 late toxic effects

I: 56%

C: 30%

p = 0.12

Neurological toxicity

I: 0/27 (0%)

C: 0/17 (0%)

p = 1.00

Taste

I: 5/27 (19%)

C: 1/17 (6%)

p > 0.05

Hearing

I: 0/27 (0%)

C: 1/17 (6%)

p > 0.05

Mandibula

I: 2/27 (6%)

C: 0/17 (0%)

p > 0.05

Teeth

I: 1/27 (4%)

C: 2/17 (12%)

p > 0.05

Xerostomia

I: 4/27 (15%)

C: 3/17 (18%)

p > 0.05

Skin and subcutaneous tissue

I: 2/27 (7%)

C: 1/17 (6%)

p > 0.05

Mucosa

I: 4/27 (15%)

C: 3/17 (18%)

p > 0.05

Definitions:

* Lateral field doses were prescribed at midplane. A supraclavicular field dose was prescribed at a 3-cm depth. If there were no palpable lymph nodes, 44 Gy was delivered in the lower part of the neck and in the spinal lymph nodes, and 56 Gy was delivered in the cervical areas adjacent to an involved lymph node area. Electron beams were used to give a boost to the posterior cervical lymph nodes. The dose to the spinal cord was kept below 44 Gy. Computed tomography scan dosimetry was performed to evaluate the maximal and minimal tumor doses.

Remarks:

-

Authors conclusion:

Calais (1999): The statistically significant improvement in overall survival that was obtained supports the use of concomitant chemotherapy as an adjunct to radiotherapy in the management of carcinoma of the oropharynx.

Denis (2004): Concomitant radiochemotherapy improved overall survival and locoregional control rates and does not statistically increase severe late morbidity. Anemia was the most important prognostic factor for survival in both arms.

Chitapanarux (2013)

Type of study:

Single-center, randomized phase III trial

Setting and country:

Thailand

Source of funding:

Not reported.

Conflicts of interest:

Not reported.

Patients with locoregionally advanced squamous-cell carcinoma of the head and neck

Inclusion criteria:

• squamous cell carcinoma of all sites of the head and neck (excluding nasopharynx, nasal cavity and paranasal sinus, and salivary gland)
• stage III or IV disease
• no distant metastases
• ECOG performance status 0-1
• age 18-75 y
• adequate laboratory organ system function
• no prior treatment

Exclusion criteria:

Not reported

N total at baseline:

Randomized: N = 102*

I: N = 48

C: N = 37

Important characteristics:

Age, median (range)

I: 58.5 y (28-70)

C: 63.5 y (40-77)

Sex, n/N (%) male

I: 38/48 (79%)

C: 55/63 (70%)

Primary tumor site

Larynx

I: 25/48 (52%)

C: 16/37 (44%)

Oropharynx

I: 8/48 (17%)

C: 6/37 (16%)

Hypopharynx

I: 2/48 (4%)

C: 2/37 (5%)

Oral cavity

I: 13/48 (27%)

C: 13/37 (35%)

Groups were comparable at baseline, except for age.

Standard fractionated radiotherapy† (total dose of 66 Gy in 6.5 wk at 2 Gy/fraction, 5 days/wk) + carboplatin (i.v. bolus of 70 mg/m² per day for 4 days on day 1-4, day 22-25 and day 43-46) and fluorouracil (600 mg/m² per day infused continuously for 4 days on day 1-4, day 22-25 and day 43-46)

Hybrid accelerated radiotherapy† (total dose of 70 Gy in 6 wk at 5 fractions of 2 Gy per wk until 40 Gy, followed by a first daily fraction of 1.8 Gy and a second daily fraction of 1.2 Gy for the remaining 30 Gy)

Length of follow-up:

Median 43 months (range: 3-102)

Loss to follow-up or missing outcome data:

None.

Clinical outcomes

Overall survival at 5 years

I: 76.1% (95%CI: 57.8-87.3)

C: 63.5 (95%CI: 42.0-78.8)

Overall survival curves were compared using the log-rank test: p = 0.05

Safety

Acute toxic effects grade ≥ 3

Skin toxicity

I: 12/48 (25.0%)

C: 6/37 (16.2%)

p = 0.33

Mucositis

I: 20/48 (41.7%)

C: 25/37 (67.6%)

p = 0.01

Anemia

I: 1/48 (2.1%)

C: 0/37 (0%)

p ≥ 0.05

Leukopenia

I: 9/48 (18.8%)

C: 0/37 (0%)

p ≥ 0.05

Thrombocytopenia

I: 3/48 (6.3%)

C: 0/37 (0%)

p ≥ 0.05

Nephrotoxicity

I: 0/48 (0%)

C: 0/37 (0%)

p = 1.00

Late toxic effects grade ≥ 3 at > 6 months

Skin

I: 3/46 (6.5%)

C: 2/25 (8.0%)

p ≥ 0.05

Subcutaneous tissue

I: 0/46 (0%)

C: 1/25 (4.0%)

p = 0.28

Mucosa

I: 1/46 (2.2%)

C: 2/25 (8.0%)

p ≥ 0.05

Salivary glands

I: 2/46 (4.3%)

C: 1/25 (4.0%)

p ≥ 0.05

Definitions:

† A shrinking field technique was used with the primary field arrangement, which included two opposed lateral fields to treat the primary tumor and upper neck nodes, and low anterior neck fields (supraclavicular fields) were treated to cover the lower neck nodes. For lateral fields, in general, the anterior border was placed at a 2-cm margin beyond the primary tumor, the posterior border was placed behind the spinous process, the superior border was placed to cover a 2-cm margin above the tumor, and the inferior border was placed at the arytenoid cartilage. The spinal cord was blocked at a dose of 40 Gy. An electron beam with appropriate energy was used to give a dose to the posterior neck field bilaterally. The low anterior low-neck field was treated with conventional fractionation to a dose of 50 Gy in both arms. All patients were treated with a 6-MV linear accelerator. Neither 3D conformal radiotherapy nor intensity-modulated radiation therapy were used in this study.

Remarks:

Due to poor accrual, the study was terminated and a final analysis was done in May 2012. We expected a 5-year OS of 25% in both treatment groups. In the overall survival analysis, in order to determine with 80% power, using a one-sided type I error of 0.20, the non-inferiority at 5-year overall survival of the two treatment arms, at least 106 patients per group had to be observed. Instead, 102 patients were enrolled and randomized.

* Efficacy was analyzed on an as-treated basis, based on the data of 85 patients. Prior to the initiation of therapy, the remaining 17 patients refused to participate in the study and were referred to other centers near their residential area.

Authors conclusion:

our results indicated that concurrent chemoradiotherapy produced a trend towards better outcome compared or hybrid accelerated radiotherapy. However, the higher hematologic toxicities and poor radiotherapy compliance rate are important limiting factors to be considered regarding this treatment.

Olmi (2003)

(ORO 93-01 trial)

Type of study:

Multicenter, randomized phase III trial

Setting and country:

18 centers of radiation therapy and medical oncology in Italy

Source of funding:

Supported in part by a grant from the Consiglio Nazionale delle Ricerche.

Conflicts of interest:

Not reported.

Patients with locoregionally advanced, unresectable epidermoid tumors of the oropharynx

Inclusion criteria:

• histologically proven epidermoid carcinoma of the oropharynx
• stage III and IV
• no distant metastases when first diagnosed
• no previous surgery (except biopsy)
• no previous RT and/or CT
• age < 70 y
• Karnofsky performance score ≥ 70 or ECOG performance status < 2
• adequate bone marrow reserve*
• serum creatinine < 1.5 mg/dL
• serum bilirubin < 1.5 mg/dL
• adequate cardiac and pulmonary function
• no previous tumors, except adequately treated in situ carcinoma of the cervix and basal cell carcinoma of the skin
• no active infectious disease
• no psychosis
• availability for correct follow-up
• informed consent obtained

Exclusion criteria:

• T1N1 and T2N1 lesions

N total at baseline:

Randomized: N = 127

I: N = 64

C: N = 63

Important characteristics:

Age,

< 50 y

I: 18/64 (28%)

C: 19/63 (30%)

50-65 y

I: 34/64 (53%)

C: 34/63 (54%)

> 65 y

I: 12/64 (19%)

C: 10/63 (16%)

Sex, n/N (%) male

I: 58/64 (91%)

C: 55/63 (87%)

Karnofsky performance score

70

I: 5/60 (8%)

C: 3/59 (5%)

80

I: 13/60 (22%)

C: 12/59 (20%)

90+

I: 42/60 (70%)

C: 44/59 (75%)

Groups were comparable at baseline.

Standard fractionated radiotherapy† (66-70 Gy in 33-35 fractions in 6.5-7 wk at 2 Gy/fraction, 5 days/wk) + carboplatin (bolus of 75 mg/m² per day infused in 30 min for 4 days in 1st, 5th and 9th wk) and fluorouracil (1000 mg/m² infused continuously in 96 hr for 4 days in 1st, 5th and 9th week)

Standard fractionated radiotherapy† (66-70 Gy in 33-35 fractions in 6.5-7 wk at 2 Gy/fraction, 5 days/wk)

Length of follow-up:

Not reported.

Loss to follow-up or missing outcome data:

None.

Clinical outcomes

Overall survival at 2 years

I: 51%

C: 40%

Overall survival curves were compared using the log-rank test: p = 0.129

Disease-free survival at 2 years

I: 42%

C: 23%

Disease-free survival curves were compared using the log-rank test: p = 0.022

HR 1.76 (95%CI: 1.15-2.91)

Safety

Acute radiotherapy-related toxic effects grade ≥ 3

Skin

I: 8/50 (16%)

C: 2/54 (4%)

p = 0.0461

Mucosa

I: 24/50 (48%)

C: 8/54 (15%)

p = 0.0003

Ear

I: 0/50 (0%)

C: 0/54 (0%)

p = 1.00

Salivary glands

I: 0/50 (0%)

C: 0/54 (0%)

p = 1.00

Pharynx

I: 12/50 (24%)

C: 6/54 (11%)

p = 0.1191

Larynx

I: 1/50 (2%)

C: 0/54 (0%)

p = 0.4808

Acute chemotherapy-related toxic effects grade ≥ 3

Nausea/vomiting

I: 0/44 (0%)

Leukopenia

I: 10/44 (23%)

Anemia

I: 1/44 (2%)

Thrombocytopenia

I: 2/44 (5%)

Alopecia

I: 0/44 (0%)

Anorexia

I: 1/44 (2%)

Nephrotoxicity

I: 1/44 (2%)

Neurotoxicity

I: 0/44 (0%)

Late toxic effects grade ≥ 3 at 2 years

Skin

I: 1/39 (3%)

C: 0/35 (0%)

p = 1.00

Subcutaneous tissue

I: 2/39 (5%)

C: 0/35 (0%)

p = 0.4946

Mucosa

I: 2/39 (5%)

C: 1/35 (3%)

p = 1.00

Salivary glands

I: 2/39 (5%)

C: 2/35 (6%)

p = 1.00

Spinal cord

I: 0/39 (0%)

C: 0/35 (0)%)

p = 1.00

Larynx

I: 0/39 (0%)

C: 1/35 (3%)

p = 0.4730

Definitions:

* Defined as a leukocyte count ≥ 4000/µL and a platelet count ≥ 150.000/µL.

† Irradiation was performed with photon beams from 60Co machines (focus-skin distance 80 cm) or 4-6-MeV linear accelerators. High-energy electron beams were used to treat the posterior regions of the neck after the spinal cord dose tolerance level was reached.  The primary tumor and both cervical regions were included in the initial treatment volume and irradiated through parallel-opposed fields (for each treatment session). The supraclavicular area was irradiated with a centrally shielded anterior direct field in case of nodal involvement of the neck.

Remarks:

The trial was officially closed in June 1998 before reaching the planned accrual of 260 patients; because the accrual rate had slowed down over the years, it was considered impractical to prolong the trial any further.

Authors conclusion:

The combination of simultaneous CT and RT with the regimen of this trial is better than RT alone in advanced oropharyngeal squamous-cell carcinomas, by increasing disease-free survival. This improvement, however, did not translate into an overall survival improvement, and was associated with a higher incidence of acute morbidity.

Staar (2001), Semrau (2006)

Type of study:

Multicenter, randomized phase III trial

Setting and country:

3 university centers and 2 community hospitals in Germany

Source of funding:

Supported by Deutsche Krebshilfe (German Cancer Aid).

Conflicts of interest:

Not reported.

Patients with locoregionally advanced, unresectable oro- and hypopharyngeal carcinomas

Inclusion criteria:

• histologically confirmed stage III or IV squamous cell carcinoma of the oro- or hypopharynx
• no distant metastases
• staged unresectable by head and neck surgeons
• no history of prior malignant neoplasm
• no previous chemo- and/or radiation therapy
• ECOG performance status 0-2
• leucocyte count > 3500 cells/ml
• absolute neutrophil count > 1500 cells/ml
• platelet count > 100.000 cells/ml
• serum creatinine ≤ 1.5 mg/dL
• creatinine clearance within normal range

Exclusion criteria:

Not reported.

N total at baseline:

Randomized: N = 263*

I: N = 113

C: N = 127

Important characteristics:

Age, median (range)

I: 57 (38-73)

C: 56 (28-73)

Sex, n/N (%) male

I: 96/113 (85%)

C: 108/127 (85%)

Tumor site

Oropharynx

I: 87/113 (77%)

C: 91/127 (72%)

Hypopharyx

I: 26/113 (23%)

C: 36/127 (28%)

Groups were comparable at baseline.

Hyperfractionated accelerated radiotherapy† (total dose of 69.9 Gy in 5.5 wk, using a concomitant boost regimen in the last 2.5 wk (wk 1-3: 1.8 Gy/fraction per day, 5 days/wk; wk 4-5.5: first daily fraction of 1.8 Gy and a second daily fraction of 1.5 Gy, 5 days/wk)) + carboplatin (i.v. bolus of 70 mg/m² on day 1-5 and day 29-33) and fluorouracil (600 mg/m² infused continuously on day 1-5 and day 29-33)

Hyperfractionated accelerated radiotherapy† (total dose of 69.9 Gy in 5.5 wk, using a concomitant boost regimen in the last 2.5 wk (wk 1-3: 1.8 Gy/fraction per day, 5 days/wk; wk 4-5.5: first daily fraction of 1.8 Gy and a second daily fraction of 1.5 Gy, 5 days/wk))

Length of follow-up:

Staar (2001): median 22.3 months (range: 6-53)

Semrau (2006): median 57.3 months (range: 10.7-84.4)

Loss to follow-up or missing outcome data:

None.

Clinical outcomes

Overall survival at 1 year for all tumors

I: 66% (95%CI: 57-75)

C: 60% (95%CI: 51-69)

Overall survival at 2 years for all tumors

I: 48% (95%CI: 38-58)

C: 39% (95%CI: 30-48)

Staar (2001): overall survival curves were compared using the log-rank test: p = 0.1139

Overall survival at 1 year for oropharyngeal tumors

I: 68%

C: 57%

95%CI: ± 10

Staar (2001): overall survival curves were compared using the log-rank test: p = 0.0468

Overall survival at 1 year for hypopharyngeal tumors

No statistically significant difference in survival curves.

Overall survival at 5 years for all tumors

I: 25.6% (95%CI: 15.8-35.4)

C: 15.8% (95%CI: 9.1-22.4)

Semrau (2006): overall survival curves were compared using the log-rank test: p = 0.016

Overall survival at 5 years for oropharyngeal tumors

I: 26.1% (95%CI: 14.3-37.8)

C: 13.0% (95%CI: 5.3-20.6)

Semrau (2006): overall survival curves were compared using the log-rank test: p = 0.008

Overall survival at 5 years for hypopharyngeal tumors

I: 22.2%

C: 22.2%

Semrau (2006): overall survival curves were compared using the log-rank test: p = 0.722

Safety

Acute toxic effects grade ≥ 3

Mucositis

I: 68%

C: 52%

p = 0.01

Dermatitis

I: 30%

C: 28%

Leukopenia

I: 18%

C: 0%

Thrombocytopenia

I: 5%

C: 0%

Anemia

I: 0%

C: 1%

Neurotoxicity

I: 0%

C: 0%

p = 1.00

Ototoxicity

I: 0%

C: 0%

p = 1.00

Nephrotoxicity

I: 0%

C: 0%

p = 1.00

Late toxic effects grade ≥ 3

Feeding tube

I: 51%

C: 25%

p = 0.02

Definitions:

† Radiotherapy was delivered by 4-6-MeV photons with immobilizing patients in thermoplastic masks for reproducible treatment. The total dose to the spinal cord should not exceed 45 Gy.

Remarks:

* For 240 of 263 randomized patients (91%), therapy started until the end of April 1999, the closure of patient enrollment. The remaining 23 patients (9%) did not start therapy, for the following reasons: (a) refusal (n = 7); (b) not qualified for protocol (n = 7); (c) infection of feeding tube (n = 3); (d) alcohol excess (n = 2); (e) distant metastases (n = 2); and (f) unknown (n = 2).

Authors conclusion:

Staar (2001):

With accelerated radiotherapy, the efficiency of simultaneously given chemotherapy may be not as high as expected when compared to standard fractionated radiotherapy. Oropharyngeal carcinomas showed better local-regional control after hyperfractionated accelerated radiochemotherapy versus hyperfractionated accelerated radiotherapy; hypopharyngeal carcinomas did not. Prophylactic granulocyte-colony stimulating factor resulted in an unexpected reduced local control and should be given in radiotherapy regimen only with strong hematologic indication.

Semrau (2006):

Hyperfractionated-accelerated concurrent radiochemotherapy is superior to hyperfractionated and accelerated radiotherapy alone in oropharyngeal carcinomas.

Study reference

Was the allocation sequence adequately generated?^a

Definitely yes

Probably yes

Probably no

Definitely no

Was the allocation adequately concealed?^b

Definitely yes

Probably yes

Probably no

Definitely no

Was knowledge of the allocated interventions adequately prevented?^c

Were patients blinded?

Were healthcare providers blinded?

Were data collectors blinded?

Were outcome assessors blinded?

Were data analysts blinded?

Definitely yes

Probably yes

Probably no

Definitely no

Was loss to follow-up (missing outcome data) infrequent?^d

Definitely yes

Probably yes

Probably no

Definitely no

Are reports of the study free of selective outcome reporting?^e

Definitely yes

Probably yes

Probably no

Definitely no

Was the study apparently free of other problems that could put it at a risk of bias?^f

Definitely yes

Probably yes

Probably no

Definitely no

Overall risk of bias

If applicable/necessary, per outcome measure^g

LOW

SOME CONCERNS

HIGH

1. Radiotherapy + cetuximab

Bonner (2006, 2010), Curran (2007)

Definitely yes

An interactive voice response system (IVRS) enabled individual sites to obtain a randomisation number and treatment group for a given patient. This process was done by PharmaNet medical monitors. Randomisation was done using a minimisation technique, which used an adaptive balancing algorithm. Patients were stratified according to their Karnofsky performance score, nodal involvement, tumour stage, and radiotherapy fractionation.

Probably yes

Authorised individuals at each institution called the IVRS system (using a pass code) and responded to certain prompts. The system then faxed the institution with details of the assigned treatment group. However, it is unclear who these ‘authorised individuals’ were and who received the details of the assigned treatment group.

Unclear

The trial was not blinded because cetuximab was known to cause an acneiform rash. After discussions with the FDA, it was felt that physician and patient blinding was not possible. However, it is unclear whether data collectors, outcome assessors and data analysts were blinded.

Definitely yes

The percentage of patients lost to follow-up is less than 10% and is similar between treatment groups.

Probably yes

All outcome measures described in the methods section are reported in the results section. However, predefined outcome measures cannot be verified in a published trial protocol.

Definitely no

Several authors report having been employed by and owning equity or stock options in the sponsors of this study. The study was designed by ImClone Systems – manufacturer of cetuximab – and the study chairman in collaboration with the lead investigators and was managed by ImClone Systems and Merck. ImClone Systems collected and analyzed the data.

HIGH (overall survival, progression-free survival, quality of life, adverse events)

It is unclear whether data collectors, outcome assessors and data analysts were blinded. One of the study sponsors collected and analyzed the data.

2. Radiotherapy + carboplatin

Fountzilas (2004)

Unclear

Not described.

Definitely yes

Stratified randomization was performed centrally at the HeCOG Data Office in Athens.

Unclear

Not described.

Definitely yes

The percentage of patients lost to follow-up is less than 10% and similar between treatment groups.

Probably yes

All outcome measures described in the methods section are reported in the results section. However, predefined outcome measures cannot be verified in a published trial protocol.

Probably no

The study suffered from a small study population (n = 79). Furthermore, it is unclear if the authors had any conflicts of interest.

HIGH (overall survival, adverse events)

The study has a small study population and it is unclear if the allocation sequence was adequately generated and whether data collectors, outcome assessors and data analysts were blinded. Also, conflicts of interest are not reported.

Jeremic (1997)

Unclear

Not described.

Unclear

Not described.

Unclear

Not described.

Definitely yes

The percentage of patients lost to follow-up is less than 10% and similar between treatment groups.

Probably yes

All outcome measures described in the methods section are reported in the results section. However, predefined outcome measures cannot be verified in a published trial protocol.

Probably no

The study suffered from a small study population (n = 106) and was stopped before reaching the planned accrual of 170 patients. Furthermore, it is unclear how the study was funded and if the authors had any conflicts of interest.

HIGH (overall survival, adverse events)

The study has a small study population and was prematurely stopped because the chief investigator had to leave the department. It is unclear if the allocation sequence was adequately generated, if the allocation was adequately concealed and whether data collectors, outcome assessors and data analysts were blinded. Also, it is unclear how the study was funded and if the authors had any conflicts of interest.

Ruo Redda (2010)

Unclear

Not described.

Unclear

Not described.

Unclear

Not described.

Definitely yes

No patients were lost to follow-up.

Probably yes

All outcome measures described in the methods section are reported in the results section. However, predefined outcome measures cannot be verified in a published trial protocol.

Probably no

The study suffered from a small study population (n = 164) and an imbalanced study population (i.e. higher percentage of patients with an ECOG performance status of 1 in the control group). Furthermore, it is unclear how the study was funded and if the authors had any conflicts of interest.

HIGH (overall survival, disease-free survival, adverse events)

The study has a small study population and an imbalanced study population. It is unclear if the allocation sequence was adequately generated, if the allocation was adequately concealed and whether data collectors, outcome assessors and data analysts were blinded. Also, potential conflicts of interest are not reported.

3. Radiotherapy + carboplatin and 5-FU

Bourhis (2012)

Definitely yes

Randomisation in a 1:1:1 ratio was done centrally at the biostatistics unit of the Institut Gustave Roussy. Doctors faxed a randomisation form to a data manager who did the randomisation with a computer program.

Definitely yes

Randomisation in a 1:1:1 ratio was done centrally at the biostatistics unit of the Institut Gustave Roussy. Doctors faxed a randomisation form to a data manager who did the randomisation with a computer program. To avoid deterministic minimisation and assure allocation concealment, the treatment which minimises the imbalance was assigned with a probability of 0.90

Unclear

Doctors and patients were not masked to treatment group assignment (open-label trial). It is unclear whether data collectors, outcome assessors and data analysts were blinded.

Probably yes

The percentage of patients lost to follow-up is less than 10%, but relatively higher in the group of patients receiving accelerated radiotherapy-chemotherapy.

Probably yes

All outcome measures described in the methods section are reported in the results section. However, predefined outcome measures cannot be verified in a published trial protocol.

Definitely yes

No potential source of bias related to the study design, baseline imbalance between treatment groups, or deviation from ITT analysis. The trial was not stopped early and the role of the funding source was clearly described.

LOW (overall survival, progression-free survival, adverse events)

It is unclear whether data collectors, outcome assessors and data analysts were blinded.

Calais (1999), Denis (2004)

Unclear

Not described.

Definitely yes

Patients were randomly assigned to a treatment group by a central office after their eligibility was established. Randomization was balanced by institution and clinical stage.

Unclear

Not described.

Definitely yes

No patient was lost to follow-up for assessment of survival. Three were lost to follow-up for the late toxicity assessment, but were free of local or distant recurrence (2 patients in intervention group and 1 patient in the control group).

Probably yes

All outcome measures described in the methods section are reported in the results section. However, predefined outcome measures cannot be verified in a published trial protocol.

Probably yes

Although the study suffered from an imbalanced study population (i.e. lower percentage of patients with N3 lymph nodes and a Karnofsky performance score of 80 in the control group), it is unlikely that this biased the results in favour of the intervention. In addition, it is unclear what the role of the funding source was. However, the study was funded by a governmental agency, not a pharmaceutical company.

SOME CONCERNS (overall survival, local recurrence, disease-free survival, adverse events)

It is unclear if the allocation sequence was adequately generated, and whether data collectors, outcome assessors and data analysts were blinded.

Chitapanarux (2013)

Unclear

Not described.

Unclear

Not described.

Unclear

Not described.

Definitely yes

No patients were lost to follow-up.

Probably yes

All outcome measures described in the methods section are reported in the results section. However, predefined outcome measures cannot be verified in a published trial protocol.

Definitely no

The study suffered from a small study population (n = 85), had an imbalanced study population (i.e. older patients in the control group) and was stopped before reaching the planned accrual of 106 patients. Furthermore, it is unclear how the study was funded and if the authors had any conflicts of interest.

HIGH (overall survival, adverse events)

The study has a small study population, an imbalanced study population and was prematurely stopped due to slow accrual. It is unclear if the allocation sequence was adequately generated, if the allocation was adequately concealed, and whether data collectors, outcome assessors and data analysts were blinded. Also, it is unclear how the study was funded and if the authors had any conflicts of interest.

Olmi (2003)

Unclear

Not described.

Definitely yes

Randomization was centralized after histologic confirmation and staging procedures; patients were stratified by center and clinical stage and were then randomized to one of the treatment groups.

Unclear

Not described.

Definitely yes

No patients were lost to follow-up.

Probably yes

All outcome measures described in the methods section are reported in the results section. However, predefined outcome measures cannot be verified in a published trial protocol.

Definitely no

The study suffered from a small study population (n = 127) and was stopped before reaching the planned accrual of 260 patients. Because the accrual rate had slowed down over the years, it was considered impractical to prolong the trial any further. The authors claim that there seemed to be no major loss in strength for the statistical estimates. Furthermore, it is unclear if the authors had any conflicts of interest.

HIGH (overall survival, disease-free survival, adverse events)

The study has a small study population and was prematurely stopped due to slow accrural. It is unclear if the allocation sequence was adequately generated, and whether data collectors, outcome assessors and data analysts were blinded. Also, it is unclear if the authors had any conflicts of interest.

Staar (2001), Semrau (2006)

Unclear

Not described.

Unclear

Not described.

Unclear

Not described.

Definitely yes

No patients were lost to follow-up.

Probably yes

All outcome measures described in the methods section are reported in the results section. However, predefined outcome measures cannot be verified in a published trial protocol.

Probably yes

No potential source of bias related to the study design, baseline imbalance between treatment groups, or deviation from ITT analysis. The trial was not stopped early and the role of the funding source was clearly described. However it is unclear if the authors had any conflicts of interest.

SOME CONCERNS (overall survival, adverse events)

It is unclear if the allocation sequence was adequately generated, and if the allocation was adequately concealed. However, treatment groups were comparable at baseline. It is unclear whether data collectors, outcome assessors and data analysts were blinded. Also, it is unclear if the authors had any conflicts of interest.

Study reference

Appropriate and clearly focused question?¹

Yes

No

Unclear

Comprehensive and systematic literature search?²

Yes

No

Unclear

Description of included and excluded studies?³

Yes

No

Unclear

Description of relevant characteristics of included studies?⁴

Yes

No

Unclear

Appropriate adjustment for potential confounders in observational studies?⁵

Yes

No

Unclear

Not applicable

Assessment of scientific quality of included studies?⁵

Yes

No

Unclear

Enough similarities between studies to make combining them reasonable?⁶

Yes

No

Unclear

Potential risk of publication bias taken into account?⁷

Yes

No

Unclear

Potential conflicts of interest reported?⁸

Yes

No

Unclear

Iocca (2018)

Yes

Unclear

The search strategy is not reported. However, the search seems comprehensive as it was conducted in PubMed, Embase, Cochrane Central Register of Controlled trials, Clinical-Trials.gov, and ASCO abstracts (from 1 January 2020 up to 1 September 2017). The search resulted in 4828 hits.

Yes

Yes

Not applicable (only RCTs are included)

Yes

This assessment involved all 57 RCTs included in the review, while only 4 RCTs are relevant for the current clinical question. Results are only presented at an aggregate level.

Not applicable

(only results of individual studies are reported)

Yes

Publication bias was assessed for the systematic review as a whole, but it was not possible to assess publication bias only for the relevant comparisons because there were fewer than ten studies available.

No

The authors declared that there was no funding source for this review.

The authors declared that there was no conflict of interest for this review.

The source of funding for the included studies was not reported.