Dose-limiting toxicities (DLTs) were drug-related toxicities defined as follows:

1. Absolute neutrophil count (ANC) < 0.5 × 10^9/L lasting for ≥ 7 days
2. ≥ Grade 3 neutropenic infection/febrile neutropenia
3. Grade 4 thrombocytopenia (TCP) or ≥ Grade 3 TCP with bleeding
4. Unable to start next cycle of chemotherapy due to lack of recovery to an ANC ≥ 1.5 × 10^9/L and platelet count ≥ 100 × 10^9/L
5. ≥ Grade 3 nonhematologic toxicity (nausea, vomiting, and diarrhea failing maximal medical management; fatigue lasting for > 72 hours)

Toxicities not clearly related to etoposide/carboplatin therapy were also considered for the purposes of determining DLTs.

Safety analysis set - included all enrolled participants (i.e., signed informed consent) who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib).

Part 1: Cohort 1 Trilaciclib 200 mg/m^2 included 1 participant from Part 2 and 1 participant from Part 1 240 mg/m^2 who received 200 mg/m^2.

An AE was defined as any untoward medical occurrence in a participant administered a medicinal product that did not necessarily have a causal relationship with this treatment. TEAEs were defined as any AE that started on or after the first dose of study drug and up to the last dose +30 days. SAEs were defined as any untoward medical occurrence that at any dose resulted in death, was life threatening, required inpatient hospitalization or prolongation of existing hospitalization, resulted in persistent or significant disability/incapacity or was a congenital anomaly/birth defect. Relatedness to study drug was assessed by the investigator. Related refers to those events that were Possibly, Probably, or Definitely Related. AEs with an unknown/not reported onset date were also included.

Safety analysis set - included all enrolled participants (i.e., signed informed consent) who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib).

Part 1: Cohort 1 Trilaciclib 200 mg/m^2 included 1 participant from Part 2 and 1 participant from Part 1 240 mg/m^2 who received 200 mg/m^2.

Severe (Grade 4) neutropenia was defined as at least 1 ANC value <0.5 × 10^9/L during the treatment period. Within each cycle, the duration (days) of severe neutropenia was defined as the number of days from the date of the first ANC value of <0.5 × 10^9/L observed between start of cycle and end of cycle to the date of the first ANC value ≥0.5 × 10^9/L that met the following criteria: 1) occurred after the ANC value of <0.5 × 10^9/L and 2) no other ANC values <0.5 × 10^9/L occurred between this day and end of cycle. The duration of severe neutropenia only included participants who had at least 1 severe neutropenia event in the cycle, and censoring rules were applied for unresolved severe neutropenia in a cycle. For the treatment period, the overall duration of severe neutropenia was the median value among the durations from all cycles.

Full analysis set (FAS) - included all randomized participants who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Cmax of trilaciclib in plasma was determined from individual concentration-time data by non-compartmental analysis methods. The actual sampling times in relation to dosing were used. For estimation of Cmax, a concentration that was below the limit of quantification (BLQ) was assigned a value of zero if it occurred in a profile before the first measurable concentration. If a BLQ value occurred after a measurable concentration in a profile, and was followed by a value above the lower limit of quantification, then the BLQ was treated as missing data. If a BLQ value occurred at the end of the collection interval (after the last quantifiable concentration) it was treated as missing data. If two BLQ values occurred in succession after Cmax, the profile was deemed to have terminated at the first BLQ value and any subsequent concentrations were omitted.

Pharmacokinetic (PK) analysis set - included all participants with evaluable PK profiles for both treatments and analytes.

AUC0-inf of trilaciclib in plasma was determined from individual concentration-time data by non-compartmental analysis methods. The actual sampling times in relation to dosing were used.

PK analysis set - included all participants with evaluable PK profiles for both treatments and analytes

Tmax of trilaciclib in plasma was determined from individual concentration-time data by non-compartmental analysis methods. The actual sampling times in relation to dosing were used.

PK analysis set - included all participants with evaluable PK profiles for both treatments and analytes

Cmax of etoposide and free and total carboplatin in plasma were determined from individual concentration-time data by non-compartmental analysis methods. The actual sampling times in relation to dosing were used. For estimation of Cmax, a concentration that was BLQ was assigned a value of zero if it occurred in a profile before the first measurable concentration. If a BLQ value occurred after a measurable concentration in a profile, and was followed by a value above the lower limit of quantification, then the BLQ was treated as missing data. If a BLQ value occurred at the end of the collection interval (after the last quantifiable concentration) it was treated as missing data. If two BLQ values occurred in succession after Cmax, the profile was deemed to have terminated at the first BLQ value and any subsequent concentrations were omitted.

PK analysis set - included all participants with evaluable PK profiles for both treatments and analytes

AUC0-inf of etoposide and free and total carboplatin in plasma were determined from individual concentration-time data by non-compartmental analysis methods. The actual sampling times in relation to dosing were used.

PK analysis set - included all participants with evaluable PK profiles for both treatments and analytes

Tmax of etoposide and free and total carboplatin in plasma was determined from individual concentration-time data by non-compartmental analysis methods. The actual sampling times in relation to dosing were used.

PK analysis set - included all participants with evaluable PK profiles for both treatments and analytes

Severe (Grade 4) neutropenia was defined as at least 1 ANC value <0.5 × 10^9/L during the treatment period. Within each cycle, the duration (days) of severe neutropenia was defined as the number of days from the date of the first ANC value of <0.5 × 10^9/L observed between start of cycle and end of cycle to the date of the first ANC value ≥0.5 × 10^9/L that met the following criteria: 1) occurred after the ANC value of <0.5 × 10^9/L and 2) no other ANC values <0.5 × 10^9/L occurred between this day and end of cycle. The duration of severe neutropenia only included participants who had at least 1 severe neutropenia event in the cycle, and censoring rules were applied for unresolved severe neutropenia in a cycle. For the treatment period, the overall duration of severe neutropenia was the median value among the durations from all cycles.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment. No participants had severe (Grade 4) neutropenia in the 240 mg/m^2 arm.

Severe (Grade 4) neutropenia was defined as at least 1 ANC value <0.5 × 10^9/L during the treatment period.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Each febrile neutropenia event (as defined by Common Terminology Criteria for Adverse Events [CTCAE]) was captured as an AE. The occurrence of febrile neutropenia was defined as at least 1 febrile neutropenia event during the treatment period. For the treatment period, the total number of febrile neutropenia events was the number of febrile neutropenia events with a unique start date.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Grade 3/4 neutropenia was defined as at least 1 ANC value <1.0 × 10^9/L during the treatment period. Within each cycle, duration (days) of Grade 3/4 neutropenia was defined as the number of days from the date of the first ANC value of <1.0 × 10^9/L observed between start of cycle and end of cycle to the date of the first ANC value ≥1.0 × 10^9/L that met the following criteria: 1) occurred after the ANC value of <1.0 × 10^9/L and 2) no other ANC values <1.0 × 10^9/L occurred between this day and end of cycle. The duration of Grade 3/4 neutropenia only included participants who had at least 1 Grade 3/4 neutropenia event in the cycle, and censoring rules were applied for unresolved Grade 3/4 neutropenia in a cycle. For the treatment period, the overall duration of Grade 3/4 neutropenia was the median value among the durations of Grade 3/4 neutropenia from all cycles.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Grade 3/4 neutropenia was defined as at least 1 ANC value <1.0 × 10^9/L during the treatment period.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Cycle nadir was the lowest value for ANC that occurred between start of cycle and end of cycle and was less than the cycle baseline.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Administration of G-CSF was collected with concomitant medications, which were coded using World Health Organization Drug Dictionary (WHO-DD) Version September 2017. A cycle where G-CSF was administered concurrently was identified by comparing the start and stop dates of each administration of G-CSF to the start of cycle and end of cycle. The occurrence of G-CSF administrations was defined as at least 1 cycle with G-CSF administrations during the treatment period. For the treatment period, the total number of G-CSF administrations was the number of cycles with G-CSF administrations.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Within a cycle, a RBC transfusion event was defined as either 1) an actual RBC transfusion, or 2) eligible for RBC transfusion (defined as hemoglobin <8.0 g/dL). The occurrence of RBC transfusions was defined as at least 1 cycle with RBC transfusion during the treatment period. For the treatment period, the total number of RBC transfusions was the number of cycles with RBC transfusions.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Blood samples were collected for local clinical laboratory assessment of hemoglobin levels.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Administration of ESAs was collected with concomitant medications, which were coded using WHO-DD Version September 2017. A cycle where an ESA was administered concurrently was identified by comparing the start and stop dates of each administration of an ESA to the start of cycle and end of cycle. The occurrence of ESA administration was at least 1 cycle with an ESA administration during the treatment period. For the treatment period, the total number of ESA administrations was the number of cycles with ESA administrations.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Within a cycle, a platelet transfusion event was defined as either 1) an actual platelet transfusion, or 2) eligible for platelet transfusion (defined as a platelet count ≤10 × 10^9/L). The occurrence of platelet transfusions was defined as at least 1 cycle with platelet transfusion during the treatment period. For the treatment period, the total number of platelet transfusions was the number of cycles with platelet transfusions.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Blood samples were collected for local clinical laboratory assessment of platelet count.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Blood samples were collected for local clinical laboratory assessment of lymphocyte count.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Dose reductions were not permitted for trilaciclib, per study protocol. Dose reductions for E/P were derived from changes in the protocol-specified dose on the dosing page and corresponded to the reductions for toxicity specified in the protocol. No more than 2 dose reductions of E/P in total were allowed for any participant. Simultaneous reductions in the doses of E/P were counted as 1 dose reduction. For the treatment period, the total number of dose reductions was the number of cycles where there was at least 1 dose reduction.

Safety analysis set - included all enrolled participants (i.e., signed informed consent) who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib).

Part 1: Cohort 1 Trilaciclib 200 mg/m^2 included 1 participant from Part 2 and 1 participant from Part 1 240 mg/m^2 who received 200 mg/m^2.

SAEs were defined as any untoward medical occurrence that at any dose resulted in death, was life threatening, required inpatient hospitalization or prolongation of existing hospitalization, resulted in persistent or significant disability/incapacity or was a congenital anomaly/birth defect. An infectious SAE was a serious event in the Medical Dictionary for Regulatory Activities (MedDRA) system organ class "infections and infestations" and a preferred term of anal abscess, bacteraemia, bronchitis, candida infection, chronic sinusitis, conjunctivitis, infection, influenza, nasopharyngitis, oral candidiasis, oral herpes, pharyngitis streptococcal, pneumonia, pneumonia bacterial, respiratory tract infection, sepsis, skin infection, upper respiratory tract infection, urinary tract infection, urosepsis or viral upper respiratory tract infection.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

SAEs were defined as any untoward medical occurrence that at any dose resulted in death, was life threatening, requires inpatient hospitalization or prolongation of existing hospitalization, resulted in persistent or significant disability/incapacity or was a congenital anomaly/birth defect. A pulmonary infection SAE was a serious event in the MedDRA system organ class "infections and infestations" and a preferred term of bronchitis, influenza, pneumonia, pneumonia bacterial, respiratory tract infection, upper respiratory tract infection or viral upper respiratory tract infection.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Intravenous antibiotic administration was collected with concomitant medications, which were coded using WHO-DD Version September 2017. A cycle where IV antibiotic was administered concurrently was identified by comparing the start and stop dates of each administration of IV antibiotic to the start of cycle and end of cycle. The occurrence of IV antibiotic administration was defined as at least 1 cycle with IV antibiotic administration during the treatment period. For the treatment period, the total number of IV antibiotic administrations was the number of cycles with IV antibiotic administrations.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

MAHE was a composite endpoint incorporating the measurement of several clinically meaningful aspects of myelopreservation into a single endpoint. The individual components for MAHE were hospitalization for a hematologic event, febrile neutropenia, death related to treatment, dose delay/reduction due to ANC or platelet counts, prolonged severe neutropenia (duration >5 days), RBC transfusion (actual or eligible) and platelet transfusion (actual or eligible). Time to first occurrence of a MAHE event was defined as the first time to observe an interested event among all the components, starting from the first dose date of study drug administration.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Tumor response was assessed by computed tomography (CT) or magnetic resonance imaging (MRI). Overall visit response by Response Evaluation Criteria in Solid Tumors (RECIST) v1.1 was derived programmatically using data from target lesions (TLs), non-target lesions (NTLs), & new lesions. Tumor response data were used to determine each participant's time point response & best overall response (BOR). Complete response (CR) was disappearance of all TLs, any pathological lymph nodes selected as TLs must have reduced in short axis to <10 mm. Partial response (PR) was at least a 30% decrease from baseline in the sum of diameters of TLs, as long as criteria for progressive disease (PD) were not met. PD was a ≥20% increase in the smallest sum of diameters of TLs since treatment started (including baseline) and an absolute increase of ≥5 mm. Stable disease (SD) was neither sufficient shrinkage to qualify for PR nor sufficient increase to qualify for PD.

Response evaluable analysis set included all participants in the safety analysis set who had at least 1 post-baseline tumor assessment, or clinical progression as noted by the investigator before their first post-baseline tumor scan, or who died due to disease progression before their first post-baseline tumor scan.

Tumor response was assessed by CT or MRI. Overall visit response by RECIST v1.1 was determined by BICR. Tumor response data were used to determine each participant's time point response & BOR. CR was disappearance of all TLs, any pathological lymph nodes selected as TLs must have reduced in short axis to <10 mm. PR was at least a 30% decrease from baseline in the sum of diameters of TLs, as long as criteria for PD were not met. PD was a ≥20% increase in the smallest sum of diameters of TLs since treatment started (including baseline) and an absolute increase of ≥5 mm. SD was neither sufficient shrinkage to qualify for PR nor sufficient increase to qualify for PD.

Response evaluable analysis set included all participants in the safety analysis set who had at least 1 post-baseline tumor assessment, or clinical progression as noted by the investigator before their first post-baseline tumor scan, or who died due to disease progression before their first post-baseline tumor scan.

Tumor response was assessed by CT or MRI. PFS was defined as the time (months) from date of first dose date of study drug for participants in Part 1 until date of documented disease progression or death due to any cause, whichever occurred first. More specifically, PFS was determined using all the assessment data up until the last evaluable visit prior to or on the date of i) disease progression as defined by RECIST 1.1 or by clinical criteria as determined by the investigator; or ii) withdrawal of consent; or iii) receiving subsequent anticancer therapy, whichever was earlier. For PFS determined using response data derived programmatically, either clinical progression or progression by RECIST (whichever came first) was considered. Median and inter-quartile range of PFS were calculated using the Kaplan-Meier method.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

OS was calculated as the time (months) from date of first dose of study drug for participants in Part 1 to the date of death due to any cause. Participants who did not die during the study were censored at the date last known to be alive. Participants lacking data beyond the day of first dose of study drug had their survival time censored at day of first dose of study drug. OS was not censored if a participant received other anti-tumor treatments after the study drugs. Median and inter-quartile range of OS were calculated using the Kaplan-Meier method.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Severe (Grade 4) neutropenia was defined as at least 1 ANC value <0.5 × 10^9/L during the treatment period.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Each febrile neutropenia event (as defined by CTCAE) was captured as an AE. The occurrence of febrile neutropenia was defined as at least 1 febrile neutropenia event during the treatment period. For the treatment period, the total number of febrile neutropenia events was the number of febrile neutropenia events with a unique start date.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Grade 3/4 neutropenia was defined as at least 1 ANC value <1.0 × 10^9/L during the treatment period. Within each cycle, duration (days) of Grade 3/4 neutropenia was defined as the number of days from the date of the first ANC value of <1.0 × 10^9/L observed between start of cycle and end of cycle to the date of the first ANC value ≥1.0 × 10^9/L that met the following criteria: 1) occurred after the ANC value of <1.0 × 10^9/L and 2) no other ANC values <1.0 × 10^9/L occurred between this day and end of cycle. The duration of Grade 3/4 neutropenia only included participants who had at least 1 Grade 3/4 neutropenia event in the cycle, and censoring rules were applied for unresolved Grade 3/4 neutropenia in a cycle. For the treatment period, the overall duration of Grade 3/4 neutropenia was the median value among the durations of Grade 3/4 neutropenia from all cycles.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Grade 3/4 neutropenia was defined as at least 1 ANC value <1.0 × 10^9/L during the treatment period.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Cycle nadir was the lowest value for ANC that occurred between start of cycle and end of cycle and was less than the cycle baseline.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Administration of G-CSF was collected with concomitant medications, which were coded using WHO-DD Version September 2017. A cycle where G-CSF was administered concurrently was identified bycomparing the start and stop dates of each administration of G-CSF to the start of cycle and end of cycle. The occurrence of G-CSF administrations was defined as at least 1 cycle with G-CSF administrations during the treatment period. For the treatment period, the total number of G-CSF administrations was the number of cycles with G-CSF administrations.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Within a cycle, a RBC transfusion event was defined as either 1) an actual RBC transfusion, or 2) eligible for RBC transfusion (defined as hemoglobin <8.0 g/dL). The occurrence of RBC transfusions was defined as at least 1 cycle with RBC transfusion during the treatment period. For the treatment period, the total number of RBC transfusions was the number of cycles with RBC transfusions. If a participant did not have any RBC transfusions, they were assigned a value of 0.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Blood samples were collected for local clinical laboratory assessment of hemoglobin levels.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Administration of ESAs was collected with concomitant medications, which were coded using WHO-DD Version September 2017. A cycle where an ESA was administered concurrently was identified by comparing the start and stop dates of each administration of an ESA to the start of cycle and end of cycle. The occurrence of ESA administration was at least 1 cycle with an ESA administration during the treatment period. For the treatment period, the total number of ESA administrations was the number of cycles with ESA administrations.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Within a cycle, a platelet transfusion event was defined as either 1) an actual platelet transfusion, or 2) eligible for platelet transfusion (defined as a platelet count ≤10 × 10^9/L). The occurrence of platelet transfusions was defined as at least 1 cycle with platelet transfusion during the treatment period. For the treatment period, the total number of platelet transfusions was the number of cycles with platelet transfusions.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Blood samples were collected for local clinical laboratory assessment of platelet count.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Blood samples were collected for local clinical laboratory assessment of lymphocyte count.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Dose reductions were not permitted for trilaciclib, per study protocol. Dose reductions for E/P were derived from changes in the protocol-specified dose on the dosing page and corresponded to the reductions for toxicity specified in the protocol. No more than 2 dose reductions of E/P in total were allowed for any participant. Simultaneous reductions in the doses of E/P were counted as 1 dose reduction. For the treatment period, the total number of dose reductions was the number of cycles where there was at least 1 dose reduction.

Safety analysis set - included all enrolled participants (i.e., signed informed consent) who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib).

Part 1: Cohort 1 Trilaciclib 200 mg/m^2 included 1 participant from Part 2 and 1 participant from Part 1 240 mg/m^2 who received 200 mg/m^2.

SAEs were defined as any untoward medical occurrence that at any dose resulted in death, was life threatening, required inpatient hospitalization or prolongation of existing hospitalization, resulted in persistent or significant disability/incapacity or was a congenital anomaly/birth defect. An infectious SAE was a serious event in the MedDRA system organ class "infections and infestations" and a preferred term of anal abscess, bacteraemia, bronchitis, candida infection, chronic sinusitis, conjunctivitis, infection, influenza, nasopharyngitis, oral candidiasis, oral herpes, pharyngitis streptococcal, pneumonia, pneumonia bacterial, respiratory tract infection, sepsis, skin infection, upper respiratory tract infection, urinary tract infection, urosepsis or viral upper respiratory tract infection.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

SAEs were defined as any untoward medical occurrence that at any dose resulted in death, was life threatening, requires inpatient hospitalization or prolongation of existing hospitalization, resulted in persistent or significant disability/incapacity or was a congenital anomaly/birth defect. A pulmonary infection SAE was a serious event in the MedDRA system organ class "infections and infestations" and a preferred term of bronchitis, influenza, pneumonia, pneumonia bacterial, respiratory tract infection, upper respiratory tract infection or viral upper respiratory tract infection.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Intravenous antibiotic administration was collected with concomitant medications, which were coded using WHO-DD Version September 2017. A cycle where IV antibiotic was administered concurrently was identified by comparing the start and stop dates of each administration of IV antibiotic to the start of cycle and end of cycle. The occurrence of IV antibiotic administration was defined as at least 1 cycle with IV antibiotic administration during the treatment period. For the treatment period, the total number of IV antibiotic administrations was the number of cycles with IV antibiotic administrations.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

MAHE was a composite endpoint incorporating the measurement of several clinically meaningful aspects of myelopreservation into a single endpoint. The individual components for MAHE were hospitalization for a hematologic event, febrile neutropenia, death related to treatment, dose delay/reduction due to ANC or platelet counts, prolonged severe neutropenia (duration >5 days), RBC transfusion (actual or eligible) and platelet transfusion (actual or eligible). Time to first occurrence of a MAHE event was defined as the first time to observe an interested event among all the components, starting from the first dose date of study drug administration.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

Tumor response was assessed by CT or MRI. Overall visit response by RECIST v1.1 was derived programmatically using data from TLs, NTLs, & new lesions. Tumor response data were used to determine each participant's time point response & BOR. CR was disappearance of all TLs, any pathological lymph nodes selected as TLs must have reduced in short axis to <10 mm. PR was at least a 30% decrease from baseline in the sum of diameters of TLs, as long as criteria for PD were not met. PD was a ≥20% increase in the smallest sum of diameters of TLs since treatment started (including baseline) and an absolute increase of ≥5 mm. SD was neither sufficient shrinkage to qualify for PR nor sufficient increase to qualify for PD.

Response evaluable analysis set included all participants in the safety analysis set who had at least 1 post-baseline tumor assessment, or clinical progression as noted by the investigator before their first post-baseline tumor scan, or who died due to disease progression before their first post-baseline tumor scan.

Tumor response was assessed by CT or MRI. Overall visit response by RECIST v1.1 was determined by BICR. Tumor response data were used to determine each participant's time point response & BOR. CR was disappearance of all TLs, any pathological lymph nodes selected as TLs must have reduced in short axis to <10 mm. PR was at least a 30% decrease from baseline in the sum of diameters of TLs, as long as criteria for PD were not met. PD was a ≥20% increase in the smallest sum of diameters of TLs since treatment started (including baseline) and an absolute increase of ≥5 mm. SD was neither sufficient shrinkage to qualify for PR nor sufficient increase to qualify for PD.

Response evaluable analysis set included all participants in the safety analysis set who had at least 1 post-baseline tumor assessment, or clinical progression as noted by the investigator before their first post-baseline tumor scan, or who died due to disease progression before their first post-baseline tumor scan.

Tumor response was assessed by CT or MRI. PFS was defined as the time (months) from date of first dose date of study drug for participants in Part 1 until date of documented disease progression or death due to any cause, whichever occurred first. More specifically, PFS was determined using all the assessment data up until the last evaluable visit prior to or on the date of i) disease progression as defined by RECIST 1.1 or by clinical criteria as determined by the investigator; or ii) withdrawal of consent; or iii) receiving subsequent anticancer therapy, whichever was earlier. For PFS determined using response data derived programmatically, either clinical progression or progression by RECIST (whichever came first) was considered. Median and inter-quartile range of PFS were calculated using the Kaplan-Meier method.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

OS was calculated as the time (months) from date of first dose of study drug for participants in Part 2 to the date of death due to any cause. Participants who did not die during the study were censored at the date last known to be alive. Participants lacking data beyond the day of first dose of study drug had their survival time censored at day of first dose of study drug. OS was not censored if a participant received other anti-tumor treatments after the study drugs. Median and inter-quartile range of OS were calculated using the Kaplan-Meier method.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.

In addition to deriving severe (Grade 4) neutropenia using the method described for the primary endpoints, an approach was applied that accounted for participants who did not experience a severe neutropenia event in Cycle 1. For the post-hoc analysis, severe neutropenia was defined as at least 1 ANC value <0.5 × 10^9/L during the treatment period. In Cycle 1, the duration (days) of severe neutropenia was defined as the number of days from the date of the first ANC value of <0.5 × 10^9/L observed between start of cycle and end of cycle to the date of the first ANC value ≥0.5 × 10^9/L that met the following criteria: 1) occurred after the ANC value of <0.5 × 10^9/L and 2) no other ANC values <0.5 × 10^9/L occurred between this day and end of cycle. The duration of severe neutropenia was set to 0 for participants who did not experience severe neutropenia in Cycle 1. Data from unscheduled visits and the actual assessment date (rather than visit date) were included in the derivation.

The FAS included all randomized patients who received at least 1 dose of study drug (etoposide, carboplatin, or trilaciclib). Analyses using the FAS were conducted on the basis of the assigned treatment.