Progression-free survival (PFS) is defined as the time from randomization to the date of progression or death due to any cause, whichever occurs first. 95% CI was obtained using Brookmeyer and Crowley method. Participants who are not observed to have progressed or died are censored at the date of last evaluable tumor assessment. Disease progression assessed as per RECISIST 1.1 was defined as at least a 20% increase in the sum of diameters of target lesions, taking as reference the smallest sum on study (this includes the baseline sum if that is the smallest on study). In addition to the relative increase of 20%, the sum must also demonstrate an absolute increase of at least 5 mm.

Intent-To-Treat Population includes all participants who are randomized into this study.

Overall survival is defined as the time from randomization to the date of death due to any cause.

Objective Response Rate (ORR) is defined as the percent of participants documented to have a confirmed complete response (CR) or partial response (PR). CR was defined as complete disappearance of all target lesions with the exception of nodal disease. All target nodes must decrease to normal size (short axis < 10 mm). All target lesions must be assessed. PR was defined as greater than or equal to 30% decrease under baseline of the sum of diameters of all target measurable lesions. The short diameter is used in the sum for target nodes, while the longest diameter is used in the sum for all other target lesions. All target lesions must be assessed. Disease progression was defined as 20% increase in the sum of diameters of target measurable lesions above the smallest sum observed (over baseline if no decrease in the sum is observed during therapy) with a minimum absolute increase of 5 mm.

Intent-To-Treat Population includes all participants who are randomized into this study.

Duration of Response (DOR) in months is defined as the time from date of the first documentation of objective response (CR or PR) to the first documentation of PD or to death due to any cause in the absence of documented PD. CR was defined as complete disappearance of all target lesions with the exception of nodal disease. All target nodes must decrease to normal size (short axis < 10 mm). All target lesions must be assessed. PR was defined as greater than or equal to 30% decrease under baseline of the sum of diameters of all target measurable lesions. The short diameter is used in the sum for target nodes, while the longest diameter is used in the sum for all other target lesions. All target lesions must be assessed. Disease progression was defined as 20% increase in the sum of diameters of target measurable lesions above the smallest sum observed (over baseline if no decrease in the sum is observed during therapy) with a minimum absolute increase of 5 mm.

Intent-To-Treat Population with Confirmed CR or PR. Intent-To-Treat Population includes all participants who are randomized into this study.

Treatment Emergent Adverse Events (TEAEs) are those that first occur or increase in severity on or after the first dose and not more than 28 days after the last dose, and prior to the initiation of subsequent systemic anti-cancer therapy. An Adverse Event (AE) is defined as any new untoward medical occurrence or worsening of a pre-existing medical condition in a clinical investigation participant administered study treatment and that does not necessarily have a causal relationship with this treatment.

Safety Population included all the participants who received at least one dose of the study drug. Docetaxel arm do not include adverse events after initiation of crossover Adagrasib treatment.

Blood samples were collected to assess hematology parameters. Adverse events are graded on a scale from 0 to 4 based on Common Terminology Criteria for Adverse Events (CTCAE) v5.0, with Grade 0 being normal Grade 1 being mild and asymptomatic; Grade 2 is moderate requiring minimal, local or noninvasive intervention; Grade 3 is severe or medically significant but not immediately life-threatening; Grade 4 events are usually severe enough to require hospitalization.

Safety Population included all the participants who received at least one dose of the study drug. Docetaxel arm do not include adverse events after initiation of crossover Adagrasib treatment.

Blood samples were collected to assess chemistry parameters. Adverse events are graded on a scale from 0 to 4 based on Common Terminology Criteria for Adverse Events (CTCAE) v5.0, with Grade 0 being normal Grade 1 being mild and asymptomatic; Grade 2 is moderate requiring minimal, local or noninvasive intervention; Grade 3 is severe or medically significant but not immediately life-threatening; Grade 4 events are usually severe enough to require hospitalization.

Safety Population included all the participants who received at least one dose of the study drug. Docetaxel arm do not include adverse events after initiation of crossover Adagrasib treatment.

Blood samples were collected for assessment of plasma concentration of Adagrasib. Data for participants for which the dose was reduced after receiving starting dose of 600 mg BID, based on physician decision is presented in separate arms.

The PK evaluable population is defined as patients who received at least one dose of active study drug and had at least one blood sample collected to assess PK concentrations

The Lung Cancer Symptom Scale (LCSS) is a disease measure of quality of life which evaluates six major lung cancer symptoms and their effect on overall distress and symptom severity, impact on day-to-day activities, and overall quality of life. This patient reported outcome (PRO) questionnaire assesses the following 6 symptoms items (appetite loss, fatigue, cough, dyspnea, hemoptysis, pain) and 3 summary global items (symptom distress, activity level, overall quality of life) for patients using visual analogue scales (VAS) (100 mm horizontal line) ranging from 0 (best rating) to 100 (worst rating). The LCSS average total score is sum of items 1 to 9 divided by the total number of items ((sum of items 1 to 9)/9) ranging from 0 to 100 where high score represent worst outcome. Least Square Mean and Confidence Interval are from a repeated measures model on the response variable change from baseline in LCSS average total score.

Intent-To-Treat Population includes all participants who are randomized into this study.

The Lung Cancer Symptom Scale (LCSS) is a disease measure of quality of life which evaluates six major lung cancer symptoms and their effect on overall distress and symptom severity, impact on day-to-day activities, and overall quality of life. This patient reported outcome (PRO) questionnaire for average symptom burden index score assesses the following six items (Appetite loss, fatigue, cough, shortness of breath, blood in sputum, pain) for patients using visual analogue scales (VAS) (100 mm horizontal line) ranging from 0 (best rating) to 100 (worst rating). The average symptom burden score is average of all the 6 items ranging from 0 to 100 where high score represent worst outcome. Least square mean and CI are from a repeated measures model on the response variable change from baseline in average symptom burden index score.

Intent-To-Treat Population includes all participants who are randomized into this study.

The Lung Cancer Symptom Scale (LCSS) is a disease measure of quality of life which evaluates six major lung cancer symptoms and their effect on overall distress and symptom severity, impact on day-to-day activities, and overall quality of life. This patient reported outcome (PRO) questionnaire for average 3-item global index score assesses the following 3 items (Distress/severity of symptoms from lung cancer, impact on normal activities, quality of life) for patients using visual analogue scales (VAS) (100 mm horizontal line) ranging from 0 (best rating) to 100 (worst rating). The 3 item global index score is average of all the 3 items ranging from 0 to 100 where high score represent worst outcome. LS mean and CI are from a repeated measures model on the response variable change from baseline in 3-item global index score.

Intent-To-Treat Population includes all participants who are randomized into this study.

The Visual Analogue Score (VAS) is a component of the EQ-5D-5L and assesses the patient's self-rated health using a vertical visual analogue scale where numbered 0 (the worst health you can image) to 100 (the best health you can imageine). The smallest change considered clinically meaningful, is defined as a score difference of 7 points.

Intent-To-Treat Population includes all participants who are randomized into this study. LS mean and CI are from a repeated measures model on the response variable change from baseline in visual analogue scale.

The European Quality of Life 5D-5L Scale (EQ-5D-5L) assesses general health-related quality of life. Health is defined in 5 dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Each dimension has 5 levels: no problems, slight problems, moderate problems, severe problems, and extreme problems. Responses are coded so that a '1' indicates no problem, and '5' indicates the most serious problem. The responses for the 5 dimensions are combined in a 5-digit number. The EQ-5D-5L health utility index (HUI) is assessed using the Crosswalk algorithm for France based on the individual responses to the 5 EQ-5D-5L domains ranging from -0.594 to 1.000. The smallest change considered clinically meaningful, is defined as a score difference of 0.08 points. Least square mean and CI are from a repeated measures model on the response variable change from baseline in health utility index.

Intent-To-Treat Population includes all participants who are randomized into this study.