OS was defined as the time from the date of randomization to the date of death due to any cause. OS (in months) was calculated as (date of death - randomization date +1)/30.4. For participants lacking survival data beyond the date of their last follow-up, the OS time was censored on the last date they were known to be alive. Participants lacking survival data beyond randomization had their OS times be censored at randomization. Estimates of OS and its 95% confidence interval were determined using Kaplan-Meier method.

The analysis population was intent-to-treat (ITT) population, which included all participants who were randomized, with study drug assignment designated according to initial randomization, regardless of whether participants received study drug or received a different drug from that to which they were randomized.

PFS was defined as the time from the date of randomization to the date of the first documentation of objective progression of disease (PD) or death due to any cause, whichever was earlier. Estimates of the PFS curves from the Kaplan Meier method were presented.

The analysis population was intent-to-treat (ITT) population, which included all participants who were randomized, with study drug assignment designated according to initial randomization, regardless of whether participants received study drug or received a different drug from that to which they were randomized.

OR was defined as the overall complete response (CR) or partial response (PR) according to the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1. Objective response rate was defined as the proportion of participants with best overall response (BOR) of CR or PR relative to all randomized.

The analysis population was ITT population, which included all participants who were randomized, with study drug assignment designated according to initial randomization, regardless of whether participants received study drug or received a different drug from that to which they were randomized.

CBR was defined as the overall CR, PR, or stable disease>=24 weeks according to the RECIST version 1.1. Clinical benefit response rate was defined as the proportion of participants with CR, PR, or stable disease>= 24 weeks relative to all randomized participants and randomized participants with measurable disease at baseline.

The analysis population was ITT population, which included all participants who were randomized, with study drug assignment designated according to initial randomization, regardless of whether participants received study drug or received a different drug from that to which they were randomized.

DR was defined as the time from the first documentation of objective tumor response (CR or PR) to the first documentation of disease progression or to death due to any cause, whichever occurred first. If tumor progression data included more than 1 date, the first date was used. DR was calculated as [the date response ended (ie, date of PD or death) - first CR or PR date + 1]/30.4.

DR was only calculated for the subgroup of all ITT participants with an objective tumor response.

AE was defined as any untoward medical occurrence in a clinical investigation participant administered a product or medical device, regardless of the causal relationship to study treatment. TEAEs were defined as AEs which occurred for the first time during the effective duration of treatment or AEs that increased in severity during treatment. Serious AEs (SAEs) were defined as any untoward medical occurrence at any dose that resulted in death; was life-threatening (immediate risk of death); required inpatient hospitalization or caused prolongation of existing hospitalization; resulted in persistent or significant disability/incapacity (substantial disruption of the ability to conduction normal life functions). AEs included SAEs and non-serious AEs. Causality to study treatment was determined by the investigator. Severity was graded according to National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events (CTCAE) version 4.03.

The analysis population included all participants who received at least 1 dose of study medication, with treatment assignments designated according to actual study treatment received.

The hematology, chemistry and coagulation tests were included in the laboratory examination. Hematology evaluation included hemoglobin, platelets, white blood cell, absolute neutrophils, absolute lymphocytes. Chemistry evaluation included alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase, sodium, potassium, magnesium, chloride, total calcium, total bilirubin, blood urea nitrogen (BUN) or urea, creatinine, uric acid, glucose (non-fasted), albumin, phosphorus or phosphate and hemoglobin A1c (HbA1c). Coagulation evaluation included activated partial thromboplastin time/partial thromboplastin time, international normalized ratio (INR) or prothrombin time.

The analysis population included all participants with at least 1 observation of the laboratory test while on study treatment or during lag time.

The EORTC QLQ-C30 is a 30 item questionnaire composed of 5 multi-item functional subscales (physical, role, cognitive, emotional, and social functioning), 3 multi-item symptom scales (fatigue, nausea/vomiting, and pain), a global health/quality of life (QOL) subscale, and 6 single items assessing other cancer related symptoms (dyspnea, sleep disturbance, appetite, diarrhea, constipation, and the financial impact of cancer). The questionnaire employed twenty-eight 4 point Likert scales with responses from "not at all" to "very much" and two 7 point Likert scales for global health and overall QOL. For functional and global QOL scales, higher scores represented a better level of functioning and all scores were converted to a 0 to 100 scale. For symptom oriented scales, a higher score represented more severe symptoms, and all scores were converted to a 0-100 scale. Negative changes from baseline indicate deterioration in functioning / global QoL scales and improvement in symptom scales.

The analysis population included all ITT participants, who had both baseline and at least 1 follow-up patient reported outcome (PRO) assessment before treatment discontinuation.

The EORTC QLQ-H&N35 is designed to be used together with the core QLQ-C30. The recall period for the items in the module was "the past week". Items hn1 to hn30 were scored on 4 point Likert type categorical scales ("not at all", "a little", "quite a bit", "very much"). Items hn31 to hn35 had a "no/yes" response format. The scores were transformed into 0 to 100 scales, with a high score implying a high level of symptoms.Negative changes from baseline indicate deterioration in functioning / global QoL scales and improvement in symptom scales.

The analysis population included all ITT participants, who had both baseline and at least 1 follow-up PRO assessment before treatment discontinuation.

A central test was defined as the tumor tissue-based p16 IHC test performed at a central laboratory (Ventana). The analysis of concordance between HPV status as assessed by local or central laboratory included the number and percentage of participants with p16 detected or not detected at the central laboratory, given that all local testing must have been negative for HPV in order for the patient to be eligible for the study. Initial analysis of the p16 status was based on the conventional cutoff of 70% p16-positive tumor cells to call out cases that might be considered HPV-positive. P16 expression was scored as positive if strong and diffuse nuclear and cytoplasmic staining was present in at least 70% of the tumor cells.

The analysis population included all participants treated with cetuximab in combination with placebo or palbociclib who had at least 1 baseline biomarker assessment.

Rb expression in the palbociclib and cetuximab treatment group, the relationship of the biomarker (individually) with PFS and OS were explored using graphical methods such as box plots, at baseline. The tumors of participants were Rb-positive, which was defined by Rb IHC with>=1% positive tumor cells.

The analysis population included all participants treated with cetuximab in combination with placebo or palbociclib who had at least 1 baseline biomarker assessment.

Ctrough is steady-state pre-dose concentration, which was observed directly from data. WPM-Ctrough is within-participant mean steady-state pre-dose concentration. For palbociclib, a steady-state trough was to be defined as a pre-dose plasma concentration following at least 7 consecutive days of 125 mg daily dose without dosing interruption and the time window for the PK collection was to be between 24 hr +/- 2 hr and 24 min post-dose the day prior to PK collection and no more than 1 hr post-dose on the day of PK collection.

The analysis population included all as-treated participants, who were treated with the study treatments and had at least measured plasma concentration for at least 1 analyte (palbociclib and/or cetuximab)

Ctrough is steady-state pre-dose concentration. Cendinf is steady-state end-of-infusion concentration. Ctrough and Cendinf were observed directly from data. WPM-Ctrough and WPM-Cendinf are within-participant mean steady-state pre-dose concentration and end-of-infusion concentration. Acceptance criteria for a steady-state Cendinf was defined as a PK sample that was 1) collected after at least 3 consecutive weeks of cetuximab IV infusions without interruption or prior dose reduction and 2) was collected at the end of cetuximab infusion time +/- 10% of the actual duration of the cetuximab infusion.

The analysis population included all as-treated participants, who were treated with the study treatments and had at least measured plasma concentration for at least 1 analyte (palbociclib and/or cetuximab)