Two tumour biopsies were collected 72 hours after the first MM-398 IV infusion during Cycle 1 of the MM-398 Treatment phase of the Pilot phase for determination of tumour levels of irinotecan and SN-38 (an active metabolite). The lesions selected for biopsy were based on the results of the FMX-MRI obtained on Days 1, 2 and 4 of the FMX phase, and were collected from a previously non-biopsied lesion. The first core biopsy was taken in the region of the tumour that showed the greatest signal change on either the T2 or T1 sequences, based on FMX-MRI. The second core biopsy was taken from the region that showed the least signal change based on FMX-MRI, avoiding areas of necrosis.

The efficacy evaluable population included all subjects who received at least 1 dose of MM-398.

Feasibility of FMX quantitation in tumour lesion was assessed through the acquisition of baseline (pre-FMX dose) and follow-up (post FMX dose) scans of sufficient quality to enable quantitative analysis to be performed. Quality was assessed by summarising scans as adequate for tumour evaluation or suboptimal but for which evaluation was completed for evaluation. Two FMX-MRI scans were taken on Day 1 (pre-FMX dosing) and on Day 2 (16-24 hours post dose) of the FMX phase. One MRI scan was also taken at 1-4 hours post FMX dose (Day 1 of FMX phase) and at 2 weeks post FMX dose (Day 15 of the MM-398 phase). It was possible for a subject to have 2 FMX-MRI scans for the same visit and timepoint corresponding to a scan target location. The number of MRI scan results that were assessed to be adequate or suboptimal at each timepoint are presented.

The pharmacodynamic evaluable population included all subjects who received at least 1 dose of MM-398 and who had pre-treatment FMX-MRI scan(s) and radiological scans at 8 weeks.

FMX tumour uptake was classified as 'low tumour uptake' or 'high tumour uptake', and was determined for 16 to 24-hours post-FMX dosing. The FMX uptake in a subject's lesions was classified using the median of the baseline-corrected FMX values at that timepoint across all subjects. The best radiological overall tumour response to MM-398 from the beginning to the end of the study was assessed using both the Investigator and imaging results per Response Evaluation Criteria in Solid Tumours (RECIST) version 1.1 (Non-central nervous system [CNS] assessments; Cohorts 1, 2 and 3). Tumour response was classified as a Complete Response (CR), Partial Response (PR), Stable Disease (SD) or Progressive Disease (PD). BOR is presented by tumour uptake classification at 16-24 hours post-FMX dose by cohort for the non-CNS RECIST assessment.

The pharmacodynamic evaluable population included all subjects who received at least 1 dose of MM-398 and who had pre-treatment FMX-MRI scan(s) and radiological scans at 8 weeks.

PFS was defined as the time in months from first dose of MM-398 to the date of radiologic disease progression by RECIST per Investigator assessment or death due to any cause, whichever occurred first.

The date of progression is defined as the earliest date that an overall tumour response of PD or death was recorded. For subjects who did not have a qualifying progressive disease or death, the date of censoring for PFS was the date when the last valid tumour assessment determined a lack of progression. The PFS assessed by the Investigator was analysed using the Kaplan-Meier method, and the median PFS based on non-CNS assessment is presented.

The efficacy evaluable population consists of all subjects who received at least 1 dose of MM-398.

PFS was defined as the time in months from first dose of MM-398 to the date of radiologic disease progression by modified RECIST (mRECIST) criteria (CNS disease; Cohort 3) per Investigator assessment or death due to any cause, whichever occurred first.

The date of progression is defined as the earliest date that an overall tumour response of PD or death was recorded. For subjects who did not have a qualifying progressive disease or death, the date of censoring for PFS was the date when the last valid tumour assessment determined a lack of progression. The PFS assessed by the Investigator was analysed using the Kaplan-Meier method, and the median PFS based on CNS mRECIST assessment is provided for Cohort 3.

The efficacy evaluable population consists of all subjects who received at least 1 dose of MM-398. Data is presented for Cohort 3 only which included subjects with brain metastasis, and who underwent CNS assessment using mRECIST criteria.

BOR was defined as the best response by RECIST version 1.1 (Non-CNS assessments) criteria per Investigator assessment, recorded from the first dose of MM-398 until disease progression or the start of new anti-cancer therapy and/or surgery.

Tumour response was classified as CR, PR, SD or PD. Classification of SD required at least 1 assessment of SD at least 4 weeks after starting treatment. Subjects were categorised as not evaluable if there was insufficient data for response classification. The BOR is presented for non-CNS assessments for the Pilot phase and Cohorts 1 - 3.

The efficacy evaluable population included all subjects who received at least 1 dose of MM-398.

BOR was defined as the best response by mRECIST criteria (CNS disease; Cohort 3) criteria per Investigator assessment, recorded from the first dose of MM-398 until disease progression or the start of new anti-cancer therapy and/or surgery.

Tumour response was classified as CR, PR, SD or PD. Classification of SD required at least 1 assessment of SD at least 4 weeks after starting treatment. Subjects were categorised as not evaluable if there was insufficient data for response classification. The BOR is presented for CNS assessments for Cohort 3.

The efficacy evaluable population included all subjects who received at least 1 dose of MM-398. Data is presented for Cohort 3 only which included subjects with brain metastasis, and who underwent CNS assessment using mRECIST criteria.

The ORR was defined as the percentage of subjects with a BOR of either a CR or PR relative to the total number of evaluable subjects. Subjects with insufficient data for response classification were classified as non-responders for objective response. The ORR is presented for non-CNS assessments for the Pilot phase and Cohorts 1 - 3.

The efficacy evaluable population included all subjects who received at least 1 dose of MM-398. Percentages are based on the number of subjects in the efficacy evaluable population in the corresponding phase/cohort.

The ORR was defined as the percentage of subjects with a BOR of either a CR or PR relative to the total number of evaluable subjects. Subjects with insufficient data for response classification were classified as non-responders for objective response. The ORR is presented for CNS assessments for Cohort 3.

The efficacy evaluable population included all subjects who received at least 1 dose of MM-398. Percentages are based on the number of subjects in the efficacy evaluable population in the corresponding phase/cohort. Data is presented for Cohort 3 which included subjects with brain metastasis, and who underwent CNS assessment using mRECIST criteria.

DOR was defined as the time from first documentation of response (CR or PR whichever occurred first, based on Investigator assessment using RECIST criteria) to the date of disease progression or to death due to any cause, whichever occurred first. DOR was computed for subjects who had CR or PR as the BOR. For subjects who did not have a qualifying progressive disease or death, the date of censoring was the date when the last valid tumour assessment determined a lack of progression. The median DOR is presented for non-CNS assessments for the Pilot phase and Cohorts 1-3.

The efficacy evaluable population included all subjects who received at least 1 dose of MM-398. Data is presented for the number of subjects who had a BOR classification of CR or PR.

DOR was defined as the time from first documentation of response (CR or PR whichever occurred first, based on Investigator assessment using mRECIST criteria) to the date of disease progression or to death due to any cause, whichever occurred first. DOR was computed for subjects who had CR or PR as the BOR. For subjects who did not have a qualifying progressive disease or death, the date of censoring was the date when the last valid tumour assessment determined a lack of progression. The median DOR is presented for CNS assessments for Cohort 3.

The efficacy evaluable population included all subjects who received at least 1 dose of MM-398. Data is presented for the number of subjects who had a BOR classification of CR or PR. Data is presented for Cohort 3 only which included subjects with brain metastasis, and who underwent CNS assessment using mRECIST criteria.

CBR was defined as the percentage of subjects with a BOR characterised as a CR at any time, PR at any time, or SD ≥ 24 weeks relative to the total number of evaluable subjects. The CBR is presented for non-CNS assessments.

The efficacy evaluable population included all subjects who received at least 1 dose of MM-398. Percentages are based on the number of subjects in the efficacy evaluable population in the corresponding phase/cohort.

CBR was defined as the percentage of subjects with a BOR characterised as a CR at any time, PR at any time, or SD ≥ 24 weeks relative to the total number of evaluable subjects. The CBR is presented for CNS assessment for Cohort 3.

The efficacy evaluable population included all subjects who received at least 1 dose of MM-398. Percentages are based on the number of subjects in the efficacy evaluable population in the corresponding phase/cohort. Data is presented for Cohort 3 which included subjects with brain metastasis, and who underwent CNS assessment using mRECIST criteria.

The number of subjects who experienced a TEAE reported to be related to MM-398 by the Investigator are presented for the Pilot and Expansion phases. An AE was considered treatment emergent if it began on or after the first administration of MM-398, started prior to dosing with MM-398 and increased in severity or seriousness after dosing, or started prior to dosing of MM-398 but the causality changed to 'related' after dosing.

The MM-398 safety population consisted of all subjects who received at least 1 dose of MM-398. Data was combined into 1 arm for the Expansion Phase as all subjects had BC.

In Cycle 1 only of the Pilot phase, samples were collected to determine the levels of total irinotecan (liposomal and free drug) and SN-38 (metabolite) in plasma following a dose of 70 mg/m^2 MM-398 FBE (80 mg/m^2 MM-398 salt-base equivalent [SBE]). The pharmacokinetic (PK) analysis was based on non-compartmental analysis. Any plasma concentrations below the lower limit of quantification (LLOQ) were assigned as missing/zero in the data set according to predetermined rules. The LLOQ for irinotecan was 0.140 micrograms per millilitre (mcg/mL), and for SN-38 the LLOQ for the Pilot phase was 0.441 nanograms per millilitre (ng/mL). The median tmax is presented for the Pilot phase.

The PK set included subjects who received at least 1 dose of MM-398, blood samples were collected at the predefined points (with no major protocol deviations affecting PK variables & sufficient number of plasma concentrations to estimate main PK parameters [Cmax, AUC]); analysis was within the sample stability period to estimate main PK parameters.

For the Expansion phase, samples were collected to determine the levels of total irinotecan (liposomal and free drug) and SN-38 (metabolite) in plasma and data is presented for Cycles 1 to 3 by dose received for each cycle (depending on dose titration): 35 to 70 mg/m^2 MM-398 FBE (40 to 80 mg/m^2 MM-398 SBE). The PK analysis was based on non-compartmental analysis. Any plasma concentrations below the LLOQ were assigned as missing/zero in the data set according to predetermined rules. The LLOQ for irinotecan was 0.140 mcg/mL and for SN-38 the LLOQ for the Expansion phase was 0.600 ng/mL. The median tmax is presented for dose levels in the Expansion phase for which data was collected. PK results for subjects in all cohorts of the Expansion Phase were combined for those on the same cycle and at the same dose. The total number of subjects evaluated in the PK set for the Expansion Phase was 21, with different numbers evaluated for each cycle/dose.

The PK set included subjects who received at least 1 dose of MM-398, blood samples were collected at the predefined points (with no major protocol deviations affecting PK variables & sufficient number of plasma concentrations to estimate main PK parameters [Cmax, AUC]); analysis was within the sample stability period to estimate main PK parameters.

In Cycle 1 only of the Pilot phase, samples were collected to determine the levels of total irinotecan (liposomal and free drug) in plasma following a dose of 70 mg/m^2 MM-398 FBE (80 mg/m^2 MM-398 SBE). The PK analysis was based on non-compartmental analysis. Any plasma concentrations below the LLOQ were assigned as missing/zero in the data set according to predetermined rules. The LLOQ for irinotecan was 0.140 mcg/mL. The mean Cmax is presented for the Pilot phase.

The PK set included subjects who received at least 1 dose of MM-398, blood samples were collected at the predefined points (with no major protocol deviations affecting PK variables & sufficient number of plasma concentrations to estimate main PK parameters [Cmax, AUC]); analysis was within the sample stability period to estimate main PK parameters.

In Cycle 1 only of the Pilot phase, samples were collected to determine the levels of SN-38 (metabolite) in plasma following a dose of 70 mg/m^2 MM-398 FBE (80 mg/m^2 MM-398 SBE). The PK analysis was based on non-compartmental analysis. Any plasma concentrations below the LLOQ were assigned as missing/zero in the data set according to predetermined rules. The LLOQ for SN-38 for the Pilot phase was 0.441 ng/mL. The mean Cmax is presented for the Pilot phase.

The PK set included subjects who received at least 1 dose of MM-398, blood samples were collected at the predefined points (with no major protocol deviations affecting PK variables & sufficient number of plasma concentrations to estimate main PK parameters [Cmax, AUC]); analysis was within the sample stability period to estimate main PK parameters.

For the Expansion phase, samples were collected to determine the levels of total irinotecan (liposomal and free drug) in plasma and data is presented for Cycles 1 to 3 by dose received for each cycle (depending on dose titration): 35 to 70 mg/m^2 MM-398 FBE (40 to 80 mg/m^2 MM-398 SBE). The PK analysis was based on non-compartmental analysis. Any plasma concentrations below the LLOQ were assigned as missing/zero in the data set according to predetermined rules. The LLOQ for irinotecan was 0.140 mcg/mL. The mean Cmax is presented for the Expansion phase. PK results for subjects in all cohorts of the Expansion Phase were combined for those on the same cycle and at the same dose. The total number of subjects evaluated in the PK set for the Expansion Phase was 21, with different numbers evaluated for each cycle/dose.

The PK set included subjects who received at least 1 dose of MM-398, blood samples were collected at the predefined points (with no major protocol deviations affecting PK variables & sufficient number of plasma concentrations to estimate main PK parameters [Cmax, AUC]); analysis was within the sample stability period to estimate main PK parameters.

For the Expansion phase, samples were collected to determine the levels of SN-38 (metabolite) in plasma and data is presented for Cycles 1 to 3 by dose received for each cycle (depending on dose titration): 35 to 70 mg/m^2 MM-398 FBE (40 to 80 mg/m^2 MM-398 SBE). The PK analysis was based on non-compartmental analysis. Any plasma concentrations below the LLOQ were assigned as missing/zero in the data set according to predetermined rules. The LLOQ for SN-38 for the Expansion phase was 0.600 ng/mL. The mean Cmax is presented for dose levels in the Expansion phase for which data was collected. PK results for subjects in all cohorts of the Expansion Phase were combined for those on the same cycle and at the same dose. The total number of subjects evaluated in the PK set for the Expansion Phase was 21, with different numbers evaluated for each cycle/dose.

The PK set included subjects who received at least 1 dose of MM-398, blood samples were collected at the predefined points (with no major protocol deviations affecting PK variables & sufficient number of plasma concentrations to estimate main PK parameters [Cmax, AUC]); analysis was within the sample stability period to estimate main PK parameters.

In Cycle 1 only of the Pilot phase, samples were collected to determine the levels of total irinotecan (liposomal and free drug) in plasma following a dose of 70 mg/m^2 MM-398 FBE (80 mg/m^2 MM-398 SBE). The PK analysis was based on non-compartmental analysis. Any plasma concentrations below the LLOQ were assigned as missing/zero in the data set according to predetermined rules. The LLOQ for irinotecan was 0.140 mcg/mL. The mean AUC(0-tlast) is presented for the Pilot phase.

The PK set included subjects who received at least 1 dose of MM-398, blood samples were collected at the predefined points (with no major protocol deviations affecting PK variables & sufficient number of plasma concentrations to estimate main PK parameters [Cmax, AUC]); analysis was within the sample stability period to estimate main PK parameters.

In Cycle 1 only of the Pilot phase, samples were collected to determine the levels of SN-38 (metabolite) in plasma following a dose of 70 mg/m^2 MM-398 FBE (80 mg/m^2 MM-398 SBE). The PK analysis was based on non-compartmental analysis. Any plasma concentrations below the LLOQ were assigned as missing/zero in the data set according to predetermined rules. The LLOQ for SN-38 for the Pilot phase was 0.441 ng/mL. The mean AUC(0-tlast) is presented for the Pilot phase.

The PK set included subjects who received at least 1 dose of MM-398, blood samples were collected at the predefined points (with no major protocol deviations affecting PK variables & sufficient number of plasma concentrations to estimate main PK parameters [Cmax, AUC]); analysis was within the sample stability period to estimate main PK parameters.

For the Expansion phase, samples were collected to determine the levels of total irinotecan (liposomal and free drug) in plasma and data is presented for Cycles 1 to 3 by dose received for each cycle (depending on dose titration): 35 to 70 mg/m^2 MM-398 FBE (40 to 80 mg/m^2 MM-398 SBE). The PK analysis was based on non-compartmental analysis. Any plasma concentrations below the LLOQ were assigned as missing/zero in the data set according to predetermined rules. The LLOQ for irinotecan was 0.140 mcg/mL. The mean AUC(0-tlast) is presented for the Expansion phase. PK results for subjects in all cohorts of the Expansion Phase were combined for those on the same cycle and at the same dose. The total number of subjects evaluated in the PK set for the Expansion Phase was 21, with different numbers evaluated for each cycle/dose.

The PK set included subjects who received at least 1 dose of MM-398, blood samples were collected at the predefined points (with no major protocol deviations affecting PK variables & sufficient number of plasma concentrations to estimate main PK parameters [Cmax, AUC]); analysis was within the sample stability period to estimate main PK parameters.

For the Expansion phase, samples were collected to determine the levels of SN-38 (metabolite) in plasma and data is presented for Cycles 1 to 3 by dose received for each cycle (depending on dose titration): 35 to 70 mg/m^2 MM-398 FBE (40 to 80 mg/m^2 MM-398 SBE). The PK analysis was based on non-compartmental analysis. Any plasma concentrations below the LLOQ were assigned as missing/zero in the data set according to predetermined rules. The LLOQ for SN-38 for the Expansion phase was 0.600 ng/mL. The mean AUC(0-tlast) is presented for dose levels in the Expansion phase for which data was collected. PK results for subjects in all cohorts of the Expansion Phase were combined for those on the same cycle and at the same dose. The total number of subjects evaluated in the PK set for the Expansion Phase was 21, with different numbers evaluated for each cycle/dose.

The PK set included subjects who received at least 1 dose of MM-398, blood samples were collected at the predefined points (with no major protocol deviations affecting PK variables & sufficient number of plasma concentrations to estimate main PK parameters [Cmax, AUC]); analysis was within the sample stability period to estimate main PK parameters.