Phase I study of bevacizumab and temsirolimus combination therapy in advanced malignancies: safety, efficacy, and ovarian cancer expansion

Abstract

Background

Bevacizumab and temsirolimus target angiogenic and mTOR pathways in cancer progression.

Methods

This phase I study enrolled 48 heavily pretreated patients with advanced solid tumors, including an ovarian cancer expansion cohort. Patients received bevacizumab biweekly plus temsirolimus weekly in a 3 + 3 design to assess safety, maximum tolerated dose (MTD) and dose-limiting toxicities (DLTs). Exploratory analyses included tumor genomic profiling and dynamic contrast-enhanced MRI (DCE-MRI).

Results

Patients had a median age of 59 and median four prior therapies. Common tumor types were ovarian (27%) and head and neck (15%). Treatment-related adverse events occurred in 93.8%, with 31.3% ≥grade 3. Five patients experienced DLTs, including grade 3 enteritis, fatigue, bowel obstruction/abdominal ileus/pulmonary embolism, bowel perforation and grade 3/4 elevated liver enzymes. MTD was bevacizumab 10 mg/kg biweekly plus temsirolimus 20 mg weekly. Overall, objective response rate (ORR) was 7.3% and 19.5% achieved stable disease ≥6 months (clinical benefit rate [CBR] 26.8%). In ovarian cohort, ORR was 16.7% and CBR 33.3%. Patients with tumor regression on DCE-MRI had lower ΔK^trans values.

Conclusion

Combination therapy showed acceptable safety and modest activity. Molecular and imaging findings were exploratory and limited. These preliminary observations could inform future biomarker studies. (ClinicalTrials.gov Identifier: NCT01552434)

Lessons Learned.

• Bevacizumab plus temsirolimus is feasible, with a defined maximum tolerated dose and a manageable toxicity profile in heavily pretreated patients.

• Clinical benefit was observed across multiple tumor types, with durable stable disease in neurofibromatosis type 2, melanoma, breast, ovarian, cervical and salivary gland cancers. Partial responses (PRs) in ovarian and cervical cancers.

• Exploratory molecular and imaging biomarkers were preliminary and may help inform the identification of responsive subgroups, with prospective validation required.

Discussion

This phase I study evaluated the combination of bevacizumab and temsirolimus in patients with advanced solid tumors. The maximum tolerated dose (MTD) was defined as dose level (DL) 7: bevacizumab 10 mg/kg biweekly plus temsirolimus 20 mg weekly, a 20% reduction from the FDA-approved temsirolimus dose. The regimen was feasible, with manageable but frequent toxicity. Overall, 62.5% of patients experienced grade (G) ≤ 2 treatment-related adverse events (TRAEs), while 31.3% developed G ≥ 3 toxicities. Common TRAEs included fatigue, mucositis, hypertriglyceridemia, skin rash, nausea, and hematologic abnormalities, consistent with prior reports.^1–3 Although anemia has been more frequently reported in prior studies, in our study it was less pronounced and was observed predominantly among patients with gynecologic cancers.^4–6 Treatment interruptions (66%), discontinuations (19%), and dose reductions (14.6%) highlight the burden of therapy, paralleling rates in other bevacizumab and temsirolimus combination studies. Compared with our group’s prior triplet regimen with valproic acid, which was associated with 55% G3/4 events and a 14.8% discontinuation rate (toxicities), this combination/schedule appeared more tolerable.⁷

Antitumor activity was modest. The ORR was 7.3% and CBR was 26.8%. Objective responses included ovarian cancer (PR –42%, PIK3CA amplification, TP53 Y220C), cervical cancer (PR –32%, PIK3CA E542K/I391M), and ovarian cancer (PR –32%, KRAS G12D). Durable stable disease was observed across ovarian, cervical, breast, melanoma, salivary gland, and neurofibromatosis type 2 (NF2) tumors, with treatment durations of 7–14 months (Table 1). These results are consistent with prior studies of single-agent bevacizumab, temsirolimus monotherapy, and our earlier bevacizumab/temsirolimus/valproic acid regimen.⁷ However, in these studies efficacy has shown considerable variability across tumor types, with higher activity in hepatocellular carcinoma (HCC), renal cell carcinoma (RCC), and endometrial cancer, but limited benefit in extra-pancreatic neuroendocrine tumors and glioblastoma.^4,6,8,9

Table 1.

Stable disease (SD) ≥ 6 months or partial response (PR) by RECISTv1.0 and characterization by patient.

Disease type	Dose Level	Best response by RECIST v1.0	Number of prior systemic regimens	Duration of treatment (months)	PTEN mut	PIK3CA mut	RAS mut	RAF mut	P53 mut
NF2	2	SD (+4%)	1	7	ND	ND	ND	ND	ND
Melanoma	2	SD (−10%)	4	11	ND	N	NRAS Q61R	N	ND
NF2	3	SD (0%)	0	8	ND	ND	ND	ND	ND
Adenocystic carcinoma of salivary gland	4	SD (−13%)	0	14	ND	ND	ND	ND	ND
Ovarian	4	PR (−42%)	5	6	N	Amplification	N	N	Y220C
ER+ PR+ Breast	5	SD (−13%)	9	8	N	E545K	ND	N	N
Cervical	7	PR (−32%)	2	3	N	E542K, I391M	N	N	N
Ovarian	7	PR (−32%)	4	5	N	N	KRAS G12D	N	N
Cervical	7	SD (−28%)	2	12	N	E81K	N	N	N
Ovarian	7	SD (−15%)	4	7	N	H1047R	N	N	C238S
Ovarian	7	SD (−26%)	5	13	N	N	N	N	N

Abbreviations: ID, identification; RECIST, Response Evaluation Criteria in Solid Tumors; mut, mutation; ND, not done; N, no; ER, estrogen receptor; PR+; progesterone receptor+; PR, partial response; SD, stable disease; NF2, neurofibromatosis type 2; PIK3CA, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha; PTEN, Phosphatase and tensin homolog; KRAS, Kirsten rat sarcoma viral oncogene homolog; NRAS, neuroblastoma ras viral oncogene homolog; RAF, rapidly accelerated fibrosarcoma; TP53, tumor protein P53.

Molecular analysis showed frequent PIK3CA mutations, with lower PTEN and TP53 mutations, likely due to limited sample size and assay constraints of the clinical testing platforms employed. Exploratory DCE-MRI in ovarian cancer cases suggested a possible correlation between vascular changes and tumor shrinkage, although limited numbers precluded definitive conclusions. Collectively, this study demonstrated that bevacizumab plus temsirolimus is tolerable with modest clinical activity. Given our results and the negative phase III data from bevacizumab plus temsirolimus in RCC, no further clinical development of this combination is planned.¹⁰

Trial information 

Trial information
Disease	Relapsed or refractory advanced solid tumors
Stage of disease/Treatment	IV
Prior therapy	No limit
Type of study	Phase I
Primary endpoints	Safety, maximum tolerated dose (MTD) and dose-limiting toxicities (DLTs).
Secondary endpoints	Preliminary efficacy (RECISTv.1.0) and exploratory dynamic contrast-enhanced MRI (DCE-MRI) analyses
Dose escalation design	3 + 3
This single-center, phase I study (NCT01552434) evaluated the safety, tolerability, and maximum tolerated dose (MTD) of bevacizumab plus temsirolimus in patients with advanced solid tumors. Secondary objectives included preliminary efficacy and exploratory DCE-MRI analyses. A standard 3 + 3 design was used.11 Adverse events (AEs) were graded per Common Terminology Criteria for Adverse Events, version 3.0 (CTCAE v3.0).

Drug information 

Drug information - 1
Generic/Working name	Bevacizumab
Company name	Genentech
Drug type	Antibody
Drug class	Anti-angiogenic; anti-VEGF-A
Dose	2.5-10
Unit	mg/kg
Route	IV
Schedule of administration	Biweekly, on day 1 and 15 of 28-day cycle

Drug information - 2
Generic/Working name	Temsirolimus
Company name	Wyeth Pharmaceuticals, Inc.
Drug type	Small molecule inhibitor
Drug class	Specific inhibitor of mTOR
Dose	5-25
Unit	mg
Route	IV
Schedule of administration	Weekly, on day 1, 8, 15, and 22 of 28-day cycle

Dose-escalation schedule for bevacizumab/temsirolimus 

Dose level	N	Temsirolimus IV on day 1, 8, 15, 22	Bevacizumab IV on day 1, 15
−1	0	5 mg	2.5 mg/kg
0	0	5 mg	5 mg/kg
1	3	5 mg	10 mg/kg
2	4	12.5 mg	2.5 mg/kg
3	3	12.5 mg	7.5 mg/kg
4	4	12.5 mg	10 mg/kg
5	3	20 mg	2.5 mg/kg
6	6	20 mg	7.5 mg/kg
7	3	20 mg	10 mg/kg
8	4	25 mg	2.5 mg/kg
9	0	25 mg	5 mg/kg
10	0	25 mg	10 mg/kg
MTD Expansion (DL7)	10	20 mg	10 mg/kg
Ovarian expansion	8	20 mg	10 mg/kg

Patient characteristics 

Characteristic	Total (%)
Number of patients	48
Male	16 (33)
Female	32 (67)
Age, years
Median (Range)	59 (19-85)
Number of prior systemic therapies
Median (Range)	4 (0-9)
ECOG performance status
0	11 (23)
1	32 (67)
2	5 (10)
Prior treatment
Surgery	41 (85)
Radiation	25 (52)
Chemotherapy	43 (90)
Prior Avastin	21 (44)
Prior mTOR inhibitor (everolimus)	2 (4)
Tumor Type
Ovarian	13 (27)
Head and neck	7 (15)
Neurofibromatosis type 2	4 (8)
Cervical	4 (8)
Sarcomaa	4 (8)
Colorectal	2 (4)
Breast	2 (4)
Melanoma	2 (4)
Unknown primary	2 (4)
Other	8 (17)
Reason for discontinuation
Disease progression	15 (31)
Toxicities	9 (19)
Withdrew consent	8 (17)
Patient’s choice to receive treatment locally	6 (13)
Clinical progression	5 (10)
Investigator’s discretion	2 (4)
Hospitalization	1 (2)
Study complication- fistula between stomach and tumoral mass (1) and facial cellulitis (1)	2 (4)

Abbreviation: ECOG, Eastern Cooperative Oncology Group.

^aIncludes: angiosarcoma (1), liposarcoma (1), pleomorphic liposarcoma (1) hemangiopericytoma (1). The less common disease types, e.g., prostate (1), renal (1), liver (1), esophageal (1), lung (1), meningioma (1), glioblastoma (1), ampullary (1), were grouped under “Other.”

Primary assessment method 

Primary assessment method
Title	Objective response rate (ORR)/clinical benefit rate (CBR)
Number of patients screened	69
Number of patients enrolled	48
Number of patients evaluable for toxicity	48
Number of patients evaluated for efficacy	41
Evaluation method	Best response by Response Evaluation Criteria in Solid Tumors version 1.0 (RECISTv1.0)
Response assessment
CR	0
PR	3 (7.3%)
SD	8 (19.5%)

Clinical benefit was defined as complete response (CR), partial response (PR) and prolonged stable disease (SD) ≥6 months. Tumor response assessment was performed by computed tomography (CT) or magnetic resonance imaging (MRI) scans at baseline and every two cycles (each cycle = 28 days) thereafter. Measurable target lesions were used to assess treatment response following RECISTv1.0.¹²

Outcome notes: 

Patient characteristics 

This study was conducted at MD Anderson Cancer Center (MDACC) from March 2012 to July 2020. Forty-eight patients were enrolled and treated with the combination therapy of bevacizumab and temsirolimus. Baseline demographic and clinical characteristics of the patients are summarized in the patient characteristics table. Thirty-two (67%) patients were female, with a median age of 59 years (range, 19-85). Most (90%) had an Eastern Cooperative Oncology Group Performance Status (ECOG) performance score of 1 or less. A total of 43 (90%) patients had received prior systemic therapy, with a median of 4 prior lines (range 0-9). No patients had prior temsirolimus; however, 2 (4%) had received everolimus and 21 (44%) had received prior bevacizumab. The most common disease types were ovarian (27%), head and neck (15%), NF2 tumor—schwannoma (8%), cervical (8%), and sarcoma (8%). The study treatment cycle consisted of 28 days, and the median number of cycles received was 3 (range, 1-14). All patients discontinued from trial, primarily due to disease progression (31%), toxicities (19%), withdrawal of consent (17%), local treatment preference (13%), clinical progression (10%), investigator’s decision (4%), hospitalization (2%), and individual complications (gastric fistula or facial cellulitis, 2% each).

Toxicity assessment 

Forty-eight patients received bevacizumab plus temsirolimus and were evaluable for safety. A total of 354 TRAEs were reported; 45 patients (93.8%) experienced ≥1 TRAE. Frequent TRAEs included fatigue (66.7%), mucositis (45.8%), hypertriglyceridemia (39.6%), and skin rash (39.6%). Grade (G) ≥3 TRAEs occurred in 15 patients (31.3%) including dose-limiting toxicities (DLTs) such as enteritis, fatigue, abdominal obstruction/pulmonary embolism, elevated aspartate aminotransferase (AST)/elevated alanine aminotransferase (ALT) and bowel perforation. G4 TRAEs were thrombocytopenia and elevated AST. No deaths were reported on study. During treatment, 66% (n = 32) of patients experienced treatment interruptions, 19% (n = 9) discontinued due to toxicity and 14.6% (n = 7) required dose modifications; of these, 42.9% (3/7) occurred during the first cycle.

Eight dose levels were evaluated, with 5 patients experiencing DLTs. The first DLT (G3 enteritis) occurred at DL6. At DL8, 2 of 4 patients developed DLTs: G3 abdominal obstruction/ileus with G3 pulmonary embolism, and G3 fatigue. DL7 was declared as the MTD. Per protocol, 10 additional patients were treated at DL7, where 2 additional DLT events occurred (G3/4 liver enzymes and G3 bowel perforation).

Antitumor activity 

Of 48 treated patients, 41 (85.4%) were response-evaluable (Figure 1A). Seven patients without post-treatment tumor reassessment were excluded from efficacy analysis. The median study duration was 3.45 months (95% confidence interval: 2.20-4.40 months). At data cutoff (August 1, 2023), all patients had discontinued treatment (Figure 1B).

Among 41 evaluable patients, 11 (26.8%) derived clinical benefit (PR + SD ≥6 months), including 3 PRs (7.3%), and 8 patients (19.5%) with SD for ≥6 months.

In total, 17 patients with gynecological cancers were enrolled (13 ovarian and 4 cervical); 15 (88.2%) were evaluable as two patients discontinued early due to treatment-related toxicity or G3 bowel perforation. All gynecological cancer patients were mTOR inhibitor-naïve and 52.9% (n = 9) had received prior bevacizumab. Among evaluable gynecological patients, 73.3% (11/15) had tumor shrinkage (median −15%, range: +31% to −42%), with an ORR of 20% and CBR of 40%. In the ovarian subgroup, 66.7% (8/12) achieved tumor reduction (ORR 16.7%, CBR 33.3%).

OncoPrint analysis and treatment response 

Available archival tumor samples were analyzed to identify mutations in PIK3CA, PTEN, RAF, RAS, and TP53 (Table 2 and Figure 1C). PIK3CA mutations were the most frequent (44.4%, 16/36) followed by TP53 mutations (32%,10/31), RAS mutations (24.3%, 9/37), PTEN (9.1%, 3/33), and RAF (2.8%, 1/36). The most common PIK3CA alteration was E542K (6/36) patients. In molecularly defined subgroups, patients with PIK3CA alterations had an ORR of 12.5% (2/16) and CBR of 31.3% (5/16). No responses were observed in PTEN-mutated tumors (0/3) and RAF-mutated tumors (0/1). Among RAS-mutated tumors, ORR was 11.1% (1/9) and CBR 22.2% (2/9). For TP53 alterations, ORR was 10% (1/10) and CBR 20% (2/10) (Table 3).

DCE-MRI 

Baseline DCE-MRI was acquired on 8 patients in the ovarian cancer expansion cohort. Two patients were excluded from DCE-MRI analysis due to missing post-treatment scans at the second time point or image artifact that precluded baseline K_tran quantification. The absolute change in the mean K^trans (ΔK^trans) and corresponding reported best RECISTv1.0 response for each of the six assessed patients can be found in Figure 2A and B. Simple linear fit to this data gives mean ΔK^trans = −2.58 × best RECIST response + 0.664 with an R² of 0.124. Subgroups with tumor regression (−26% to −10%) had lower ΔK^trans values (−0.44 to 0.47 min−¹) while a higher ΔK^trans (−0.14 to 3.06 min−¹) were observed in patients with positive tumor growth (2% to 21%). However, the limited sample size precludes drawing definitive conclusions.

Assessment, analysis and discussion
Completion	Study completed
Investigator’s assessment	Level of activity is modest, and no further study planned

≥Grade 3 adverse events and dose-limiting toxicities across bevacizumab and temsirolimus combination therapy dose levels 

Dose level	n	Temsirolimus IV on day 1, 8, 15, 22	Bevacizumab IV on day 1, 15	SD ≥ 6 months or PR/total treated	DLT	Grade (G) 3/4 toxicity (n)
1	3	5 mg	10 mg/kg	0/3	0/3	G3 Neutropenia (1)
2	4	12.5 mg	2.5 mg/kg	2/4	0/4
3	3	12.5 mg	7.5 mg/kg	1/3	0/3	G3 cellulitis (1)
4	4	12.5 mg	10 mg/kg	2/4	0/4	G3 mucositis (1)
5	3	20 mg	2.5 mg/kg	1/3	0/3
6	6	20 mg	7.5 mg/kg	0/6	1/6	G3 Enteritis (1)^ G4 Thrombocytopenia (1)
7	3	20 mg	10 mg/kg	2/3	0/3
8	4	25 mg	2.5 mg/kg	0/4	2/4	G3 Fatigue (1)^ G3 Hypophosphatemia (1) G3 bowel obstruction/abdominal ileus and G3 Pulmonary embolism (1)^ Epistaxis (1)
MTD expansion cohort (DL7)	10	20 mg	10 mg/kg	0/10	2/10	G3 Anorexia (1) G3 Elevated ALT/G4 elevated AST (1)^ G3: Bowel Perforation (1)^ G3 Hypophosphatemia (1) G4 Thrombocytopenia (1)
Ovarian expansion cohort	8	20 mg	10 mg/kg	3/8	0/8	G3 Leukopenia (1) G3 Hypertriglyceridemia (1) G3 Diarrhea (1) G3 Neutropenia G4 Thrombocytopenia (1)

Adverse events deemed at least possibly related to treatment, graded based on Common Terminology Criteria for Adverse Events, version 3.0 (CTCAEv3.0). Symbol: ^ indicates a dose-limiting toxicity (DLT). The denominator in the DLT column reflects the number of patients evaluable for DLT at each dose level.

Abbreviations: DL, dose level; mg, milligram; kg, kilogram; IV, intravenous; PR, partial response.

General toxicity profile 

Adverse events 

	Cohort
	1 (n = 3)		2 (n = 4)		3 (n = 3)		4 (n = 4)		5 (n = 3)		6 (n = 6)		7 (n = 13)		8 (n = 4)		Ovarian expansion (n = 8)		All (%, n = 48)
Adverse events	≤G2	≥G3	≤G2	≥G3	≤G2	≥G3	≤G2	≥G3	≤G2	≥G3	≤G2	≥G3	≤G2	≥G3	≤G2	≥G3	≤G2	≥G3	≤G2	≥G3
Bowel obstruction	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	1 (2.1)
Alter taste	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1	0	2 (4.2)	0
Anemia	0	0	0	0	0	0	0	0	0	0	0	0	3	0	2	0	2	0	7 (14.6)	0
Anorexia	0	0	1	0	0	0	1	0	1	0	5	0	4	1	1	0	1	0	14 (29.2)	1 (2.1)
Abdominal pain	0	0	1	0	0	0	2	0	0	0	0	0	1	0	1	0	4	0	9 (18.8)	0
Bowel perforation	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	1 (2.1)
Burning sensation when urinating	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	1 (2.1)	0
Chest tightness	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1 (2.1)	0
Constipation	0	0	0	0	0	0	0	0	1	0	1	0	3	0	0	0	0	0	5 (10.4)	0
Cough	0	0	0	0	0	0	0	0	1	0	0	0	1	0	0	0	1	0	3 (6.3)	0
Dry skin	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	1 (2.1)	0
Dry cracking hands	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	1 (2.1)	0
Decreased sodium	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	1 (2.1)	0
Dry mouth	0	0	1	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	2 (4.2)	0
Diabetes Mellitus type II	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	1 (2.1)	0
Dizziness	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1 (2.1)	0
Dyspnea	0	0	0	0	0	0	0	0	1	0	1	0	0	0	0	0	2	0	4 (8.3)	0
Diarrhea	0	0	0	0	0	0	1	0	0	0	1	0	4	0	0	0	3	1	9 (18.8)	1 (2.1)
Elevated LDH	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1 (2.1)	0
Elevated ALT	0	0	0	0	1	0	1	0	0	0	0	0	3	1	0	0	2	0	7 (14.6)	1 (2.1)
Epistaxis	1	0	1	0	0	0	2	0	0	0	1	0	1	0	1	1	3	0	10 (20.8)	1 (2.1)
Enteritis	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	1 (2.1)
Elevated AST	0	0	0	0	1	0	1	0	0	0	0	0	5	1	2	0	3	0	12 (25)	1 (2.1)
Edema	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1 (2.1)	0
Failure to thrive	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	1 (2.1)	0
Fever	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1 (2.1)	0
Fatigue	2	0	2	0	1	0	3	0	2	0	6	0	10	0	2	1	4	0	32 (66.7)	1 (2.1)
Hypoalbuminemia	0	0	0	0	0	0	0	0	0	0	0	0	2	0	0	0	0	0	2 (4.2)	0
Hemorrhoid	0	0	0	0	0	0	0	0	0	0	0	0	2	0	0	0	0	0	2 (4.2)	0
Hypophosphatemia	0	0	0	0	0	0	0	0	1	0	0	0	1	1	1	1	0	0	3 (6.3)	2 (4.2)
Hypertriglycerides	2	0	1	0	0	0	2	0	0	0	2	0	8	0	2	0	2	1	19 (39.6)	1 (2.1)
Hypercholesterolemia	1	0	1	0	0	0	2	0	0	0	1	0	6	0	1	0	1	0	13 (27.1)	0
Hyperglycemia	0	0	0	0	0	0	0	0	0	0	2	0	3	0	2	0	2	0	9 (18.8)	0
Hyperbilirubinemia	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1	0	0	0	2 (4.2)	0
Hypertension	0	0	0	0	0	0	1	0	1	0	2	0	2	0	0	0	2	0	8 (16.7)	0
Hemoptysis	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	1	0	2 (4.2)	0
Headache	0	0	0	0	1	0	1	0	0	0	2	0	3	0	1	0	3	0	11 (22.9)	0
Hypomagnesaemia	0	0	0	0	0	0	1	0	0	0	0	0	1	0	0	0	0	0	2 (4.2)	0
Hypokalemia	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	2	0	3 (6.3)	0
Insomnia	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1 (2.1)	0
Increased creatinine	0	0	0	0	0	0	1	0	0	0	0	0	2	0	0	0	0	0	3 (6.3)	0
Increased Alkaline Phosphatase	0	0	0	0	0	0	0	0	0	0	0	0	4	0	0	0	1	0	5 (10.4)	0
Increased calcium	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	1 (2.1)	0
Increased potassium	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	1 (2.1)	0
Leukopenia	2	0	0	0	0	0	0	0	0	0	0	0	5	0	3	0	0	1	10 (20.8)	1 (2.1)
Jaw Pain	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1 (2.1)	0
Myalgia	0	0	0	0	0	0	0	0	0	0	2	0	0	0	0	0	2	0	4 (8.3)	0
Mucositis of anus	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	1 (2.1)	0
Muscle tightness	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	1 (2.1)	0
Mucositis	0	0	2	0	1	0	1	1	2	0	4	0	7	0	1	0	4	0	22 (45.8)	1 (2.1)
Nausea	1	0	1	0	0	0	1	0	2	0	3	0	4	0	1	0	3	0	16 (33.3)	0
Neutropenia	0	1	0	0	0	0	0	0	0	0	1	0	3	0	0	0	0	1	4 (8.3)	2 (4.2)
Neuropathy	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	1	0	2 (4.2)	0
Pneumonitis	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1 (2.1)	0
Pulmonary embolism	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	1	0	1 (2.1)	1 (2.1)
Pruritus	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1 (2.1)	0
Proteinuria	1	0	1	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	3 (6.3)	0
Rhinorrhea	0	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	1 (2.1)	0
Skin necrosis	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	1 (2.1)
Skin rash	0	0	2	0	1	0	2	0	1	0	3	0	5	0	1	0	4	0	19 (39.6)	0
Skin lesions	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	0	1 (2.1)	0
Sore throat	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1 (2.1)	0
Thrombocytopenia	2	0	0	0	0	0	1	0	0	0	3	1	5	1	3	0	1	1	15 (31.3)	3 (6.3)
Tightness in back	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	1 (2.1)	0
Tooth pain	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	1 (2.1)	0
UTI	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	1	0	2 (4.2)	0
Ulcer ostomy	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1 (2.1)	0
Umbilical metastasis	0	0	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	0	1 (2.1)	0
Vomiting	0	0	0	0	0	0	0	0	0	0	0	0	2	0	0	0	1	0	3 (6.3)	0
Weakness	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1 (2.1)	0
Weight loss	0	0	0	0	0	0	1	0	0	0	0	0	1	0	0	0	1	0	3 (6.3)	0
Yeast infection	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0	1 (2.1)	0

Abbreviations: UTI, urinary tract infection; AST, aspartate aminotransferase; ALT, alanine aminotransferase; LDH, lactate dehydrogenase; n, number; G, grade.

Additional details of study design

Dose-limiting toxicities (DLTs) were defined as significant TRAEs during cycle 1 (28 days). If ≥33% of patients in a cohort experienced DLTs, dose escalation ceased, and the preceding dose was defined as the MTD. A cohort could expand to 14 patients at the MTD if a meaningful response—PR, ≥25% biomarker decline, or stable disease ≥4 months—was observed.¹³ Following a PR in an ovarian cancer patient, this tumor type was selected for cohort expansion and exploratory analyses using DCE-MRI.

The study protocol and amendments were approved by The Institutional Review Board (IRB) and conducted in accordance with the Declaration of Helsinki, Good Clinical Practice and met all federal, state and local regulatory guidelines. Written informed consent was obtained from all patients. This study was investigator initiated, and drugs were obtained commercially.

Eligible patients had advanced or metastatic cancers progressing after standard therapy, Eastern Cooperative Oncology Group (ECOG) performance status ≤2, and adequate hematologic, renal, and hepatic function. Key exclusions included uncontrolled hypertension, recent major surgery (≤6 weeks), significant bleeding, pregnancy, or hypersensitivity to study drugs. Prior bevacizumab or temsirolimus treatment was allowed.

For patients with available archival tissue, molecular profiling was performed to assess alterations in key cancer-related genes including phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA), Phosphatase and Tensin homolog deleted on chromosome 10 (PTEN), rapidly accelerated fibrosarcoma (RAF), rat sarcoma (RAS), and tumor protein 53 (TP53). Mutation data were integrated and visualized using the MDACC OncoPrint platform (MCPlotter), curated and annotated through the Precision Oncology Decision Support System (PODSS).¹⁴

Baseline DCE-MRI was performed in the ovarian cancer expansion cohort within 7 days before treatment, at 48hr (±6 hr) post-treatment and at cycle 1, day 28. Eight patients underwent DCE-MRI imaging. Patients were positioned feet-first supine and imaged using posterior spine coil array used in conjunction with a flexible 18-channel anterior array, under free breathing. All images were acquired on a 3 T MRI MAGNETOM Prisma scanner (Siemens Healthcare, Erlangen, Germany) with 64-channel configuration and employed sagittal, fat-saturated 3D T1-weighted gradient-recalled echo acquisition targeting solid tumor components after IV Gd-DTPA (0.2 mmol/kg, 5 mL/s) followed by 20 ml saline. Parameters: TR/TE =3.7/1.5 ms, flip angle = 25°, matrix = 256 × 128, voxel = 1.6 × 1.6 × 5 mm. Pre-contrast T1-maps were acquired with flip angles 2°, 5°, 10°, 15°. Each volume required ∼6 s; total scan 5–6 min.

Volume transfer coefficient (K^trans) pixel-by-pixel maps were generated from the DCE-MRI using an extended Toft’s model in a commercial software package (NordicICE; NordicNeuroLab, version 2.3.14, Bergen, Norway). The T₁ value of blood was set to 1600 ms and the arterial input function was obtained from the abdominal aorta. The mean value of K^trans was computed over the center slice of the target lesions. DCE and T₁ map source images were reviewed and lesions with noticeable movement or artifacts impacting the measurements were excluded from analysis.

Extended online discussion

Bevacizumab plus temsirolimus is tolerable, with MTD established as bevacizumab 10 mg/kg biweekly plus temsirolimus 20 mg weekly- 20% lower than FDA-approved temsirolimus dose. This dose was used for exploratory DCE-MRI analysis in ovarian cancer. The regimen demonstrated manageable safety profile: 62.5% (n = 30) had G ≤ 2 TRAEs and 31.3% (n = 15) experiencing G ≥ 3 TRAEs, consistent with prior studies.^4,5,7,8 Common TRAEs included fatigue, mucositis, hypertriglyceridemia and cytopenias, while anemia was less frequent (vs 56%-71% in previous studies) and occurred primarily in patients with gynecologic cancers.^4–6,8

The median treatment duration was 3.45 months; 66% (n = 32) experienced treatment interruptions, 19% (n = 9) discontinued due to toxicity and 14.6% (n = 7) required dose modifications; of these, 42.9% (3/7) occurred during the first cycle. Comparable or higher discontinuation rates have been reported in endometrial cancer (38.8%), extra-pancreatic neuroendocrine tumors (30%) and RCC (42%).^8,15,16 Our prior gynecologic study showed fewer G ≥ 3 TRAEs, whereas our subsequent study combining bevacizumab, temsirolimus and valproic acid led to more frequent G3/4 AEs (55%) and a 14.8% discontinuation rate, reflecting added toxicity of the triplet regimen.^1,7

The combination achieved an ORR of 7.3% and a CBR of 26.8%, comparable to bevacizumab monotherapy (ORR 6.7%, CBR ∼22% in breast cancer),² temsirolimus monotherapy (ORR 6.3%-8.3%, CBR 8.3%-9.5% in advanced solid tumors),^3,17 and our prior triplet regimen with valproic acid (ORR 7.9%, CBR 21%).⁷ Activity varied by tumor type, with higher efficacy observed in hepatocellular carcinoma (ORR 19%, CBR 80.8%),⁵ RCC (ORR 28.8%, CBR 84.6%),⁴ endometrial cancer (ORR 24.5%, CBR 71.3%),⁸ and gynecologic cancers (ORR 17%, CBR 37%)¹ (Table 4). In gynecologic subgroup, CBR was 40% aligning with prior studies but lower than single-agent bevacizumab in ovarian cancer (ORR 67%, CBR 96%).¹⁸

Table 4.

Grade ≤2 treatment-related adverse events (TRAEs) with bevacizumab or temsirolimus and in combination.

	Bevacizumab	Temsirolimus	Bevacizumab + Temsirolimus
	Breast (Cobleigh et al.)	Advanced cancer (Raymond et al.)	Hepatocellular carcinoma (Knox et al.)	Renal carcinoma (Merchan et al.)	Extra-pancreatic neuroendocrine (Abuzakhm et al.)
Frequent treatment-related adverse events occurring ≥20%
Non-hematologic AE	Asthenia (57%) Nausea (51%) Infection (40%) Pain (39%) Diarrhea (37%) Vomiting (37%) Headache (36%) Myalgia (32%) Arthralgia (31%) Cough increased (28%) Dyspnea (28%) Pruritus (23%) Hypertension (23%) Sinusitis (20%) Hyperesthesia (20%) Paresthesia (20%)	Dermatologic (71%) Mucositis (71%) Asthenia (38%) Nausea (42%) Anorexia (21%) Diarrhea (17%) Vomiting (21%) Peripheral edema (21%) Weight loss (21%) Taste pervasion (21%)	Rash (60%) Fatigue (56%) proteinuria (56%) Mucositis (41%) Nausea (41%) Diarrhea (30%) Anorexia (26%) Minor Bleeds (26%) Increased cholesterol (26%) Increased triglycerides (22%) Vomiting (22%) Weight loss (22%)	Mucositis (67%) Fatigue (65%) Hypertriglyceridemia (63%) Hypercholesterolemia (63%) Proteinuria (48%) Epistaxis (42%) Rash (42%) Diarrhea (40%) Creatinine elevation (38%) Anorexia (37%) Hypertension (27%) Nausea (25%)	Hypercholesterolemia (53%) Proteinuria (50%) Fatigue (50%) Hypertension (44%) Hypertriglyceridemia (41%) Diarrhea (41%) Anorexia (39%) Mucositis (37%) Rash (35%) Epistaxis (33%) Nausea (33%) Hyperglycemia (35%) AST elevation (30%) Elevated creatinine (26%) Weight loss (26%) ALT elevation (24%) ALP elevation (24%) Hypokalemia (20%)
Hematologic AE		Thrombocytopenia (21%)	Thrombocytopenia (60%) Anemia (56%) Neutropenia (56%) Leucopenia (52%)	Anemia (71%) Thrombocytopenia (40%)	Thrombocytopenia (63%) Anemia (59%) Leukopenia (52%) Neutropenia (28%)
Response	N = 75	N = 24	N = 26	N = 52	N = 48
ORR	5 (6.7%)	2 (8.3%)	5 (19%)	15 (28.8%)	1 (2%)
SD	12 (16%)	-	16 (61.5%)	29 (55.8%)	42 (88%)
CBR	22.6%	8.3%	80.8%	84.6%	89%

Abbreviations: AST, aspartate aminotransferase; ALT, alanine aminotransferase; ALP, alkaline phosphatase; AE, adverse event; SD, stable disease; ORR, objective response rate; CBR, clinical benefit rate.

Table 2.

Tumor molecular analysis.

Gene	N/Total tested	Molecular aberration	Cancer type	Duration of response (months)	Best RECIST response
PIK3CA	36a/48	H1047R	ER+ Breast	4.3	SD (−20%)
	16b/36 (44%)	Amplification	Ovarian	5.6	PR (−42%)
		E542K	Parotid salivary ductal carcinoma	2.6	NMD
		E545K	ER + PR + Breast	7.7	SD (−13%)
		I391M	HNSCC	0.7	SD (+14%)
		E542K, I391M	Cervical	3	PR (−32%)
		E542K	Colorectal	1.0	SD (−16%)
		E542K	Cholangiocarcinoma	0.3	PD (+56%)
		E81K	Cervical	12.4	SD (−28%)
		E545G	Non-small cell lung cancer	1.9	NE
		H1047L	Ovarian	3.9	SD (−14%)
		E545K	Cervical	3.7	SD (−14%)
		E542K	HNSCC	4.4	SD (−16%)
		E542K	Cervical	0.6	NE
		K111E	Ovarian	5.4	SD (−10%)
		H1047R	Ovarian	6.8	SD (−15%)
PTEN	33a/48	P248fs*5	Glioblastoma	1.2	CPD (−3%)
	3b/33 (9.1%)	PTEN Q245  *	Cervical	3.7	SD (−14%)
		PTEN loss
		PTEN T319fs  *  1	Parotid high-grade ductal cancer	0.9	NE
RAS	37a/48	NRAS Q61R	Melanoma	10.5	SD (−10%)
	9b/37 (24.3%)	HRAS Q61R	Parotid salivary ductal carcinoma	2.6	NMD
		NRAS Q61R	Melanoma	2.2	SD (−5%)
		NRAS E153del	Neurofibromatosis type 2	5.6	SD (−18%)
		KRAS G12D	Ovarian	3.7	SD (−29%)
		KRAS Q61L	Ampullary adenocarcinoma	0.9	NE
		KRAS G12S	Colorectal	1.0	SD (−16%)
		KRAS G12D	Ovarian	5.3	PR (−32%)
		KRAS L19F	Angiosarcoma	4.6	SD (−11%)
RAF	36a/48	BRAF_V600E	Ovarian	4.8	SD (+2%)
	1b/36 (2.8%)
TP53	31a/48	Y220C	Ovarian	5.6	PR (−42%)
	10b/31 (32%)	H179D	HNSCC	0.7	SD (+14%)
		Y220C	Ovarian	3.2	SD (0%)
		T155N	Ovarian	3.9	SD (−14%)
		M169fs  *  10	Ovarian	0.3	CPD
		C238S	Ovarian	6.8	SD (−15%)
		S215R	Ovarian	1.8	SD (−26%)
		Y126C	Ovarian	4.8	SD (+2%)
		G266R	Parotid gland	0.9	NE
		R273H	Ovarian	0.5	PD (+30.7%)

Abbreviations: N, number; NE, no response evaluation; NMD no measurable disease; SD stable disease; PR, partial response; PD, progressive disease; PIK3CA, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha; PTEN, Phosphatase and tensin homolog; KRAS, Kirsten rat sarcoma viral oncogene homolog; NRAS, neuroblastoma rat sarcoma viral oncogene homolog; HRAS, Harvey rat sarcoma viral oncogene homolog; RAF, rapidly accelerated fibrosarcoma; TP53, tumor protein P53.

^aNumber of patients tested for indicated gene.

^bPositive for mutation.

Table 3.

Treatment response by molecular subgroup.

Gene	Number of patients with mutation/number of patients tested	Objective response rate (ORR)	Clinical benefit rate (CR + PR + SD ≥ 6 months)
PIK3CA	16/36 (44%)	2/16 (12.5%)	5/16 (31.3%)
PTEN	3/33 (9.1%)	0	0
RAS	9/37 (24.3%)	1/9 (11.1%)	2/9 (22.2%)
RAF	1/36 (2.8%)	0	0
TP53	10/31 (32%)	1/10 (10%)	2/10 (20%)

Abbreviations: PIK3CA, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha; PTEN, Phosphatase and tensin homolog; RAS, rat sarcoma; RAF, rapidly accelerated fibrosarcoma; TP53, tumor protein P53.

Molecular profiling was performed to identify predictors of response to bevacizumab plus temsirolimus. Alterations in PIK3CA, PTEN, RAS, and TP53 were evaluated given their roles in PI3K/AKT/mTOR and VEGF pathways targeted by this combination. PIK3CA activation or PTEN alteration enhances mTOR signaling, while TP53 dysfunction promotes VEGF-driven angiogenesis.^19,20 Tumor shrinkage in 68.8% (11/16) of PIK3CA-mutant tumors, including 2 PRs and 3 SD ≥6 months, aligned with the mechanism of action of bevacizumab and temsirolimus, and the benefit of dual blockade of angiogenic and proliferative signaling. However, the incidence of PTEN and TP53 mutations in our study was lower than previously reported, likely due to both the small sample size and assay constraints, which varied in sensitivity and exon coverage (Table 5), underscoring the importance of comprehensive genomic profiling in future studies.

Table 5.

Summary of molecular testing platforms and exon coverage for PTEN and TP53.

TP53 Patients analyzed by indicated test (total = 35)	PTEN    Patients analyzed by indicated test (total = 33)	Assay Type	Testing platform	Methodology	Sensitivity (VAF)	Exons (Codons) Tested
2	0	Tissue based	Knight Cancer Institute/Knight Diagnostic Laboratories	Sequenom MassARRAY system-multiplex PCR amplification of gene exons of interest; primer extension reaction using TYPLEX chemistry; readout of primer extension products by mass spectrometry. Multiplex panel of 496 assays that can detect up to 643 different mutations.	10%	Not provided Only approximately 20% of potential mutations are covered in the TP53 gene.
7	7	Tissue based	MDACC 50-Gene Somatic Mutation Analysis Panel	PCR-based sequencing using a NGS-platform on genomic DNA to screen for mutations in the coding sequence of 50 genes	SNV 5%-10%	Hotspot technology PTEN 1 (1-25), 3 (55-70), 5 (99-135), 6 (165-184), 7 (212-215), 7 (231-267), 8 (282-300), 8 (312-342) TP53 2(1-20), 4 (68-113), 5 (126-138), 5-6(149-223), 7 (225-258), 8 (263-307), 10 (332-367)
3	3	Tissue based	MDACC STGA v1.0	PCR-based sequencing using a NGS-platform on genomic DNA to screen for mutations in the coding sequence of 128 genes	SNV 5%-10%	Hotspot technology PTEN 1-7 (1-267), 8 (282-300), 8-9 (303-382) TP53 2-11 (1-394),
4	4	Tissue based	MDACC Cancer Mutation Scan 46 genes	PCR-based sequencing using NGS platform on genomic DNA to screen for the frequently reported point mutations. (46-gene panel).	5%-10%	Hotspot technology PTEN 6, 68, 173, 171, 214, 242, 248, 267, 290, 317, 318, 319, 321, 323, 335 TP53 11, 95, 103, 108, 110, 132, 135, 152, 155, 157, 158, 159, 163, 173, 174, 175, 176, 179, 190, 193, 194, 195, 196, 198, 199, 204, 212, 213, 216, 220, 236,237, 241, 242, 245, 248, 249, 252, 253, 270, 272, 273, 274, 275, 278, 281, 282, 285, 290, 298, 306, 336, 342
2	2	Tissue based	MDACC Cancer Mutation - 409 genes	PCR-based sequencing using NGS-platform on genomic DNA to screen for mutations in the 409 genes.	5%-10%	Hotspot technology PTEN 1 (1-27), 2 (46-55), 4 (78-85), 5 (87-115), 5 (106-135), 6-7 (179-257), 7 (259-267), 8 (282-300), 8-9 (312-348) TP53 2-7 (1-261), 8-10 (271-344), 11 (367-394)
6	6	Tissue based	MDACC STGA-DNA 2018	PCR-based sequencing using NGS-platform on genomic DNA to screen for mutations in the coding sequence of 134 genes	5%-10%	Hotspot technology PTEN 1-9 (1-382) TP53 2-11 (1-394)
10	10	Tissue based	FoundationOne	Qualitative next-generation sequencing based in vitro diagnostic test that uses targeted high throughput hybridization-based capture technology for detection of substitutions, insertion and deletion alterations. (315 genes)	Base substitutions: at mutant allele frequency 5%-10%. Insertion/deletion (1-40 bp): at mutant allele frequency 10%-20%	Hybridization capture technology. (non-hotspot)
1	1	Liquid biopsy	Guardant360 CDx	Cell-free DNA is extracted from plasma and genomic alterations are analyzed by massively parallel sequencing of amplified target genes using the Illumin HiSeq 2500 or Illumina NextSeq 500 platform. (73 genes)	The minimum detectable mutant allele fraction (limit of detection) is dependent on the patient’s sample cell-free DNA concentration, which can vary from less than 10 to over 1000 genomic equivalents per mL of peripheral blood.	Not provided

Abbreviations: MDACC, MD Anderson Cancer Center; STGA, solid tumor genomic assay; VAF; variant allele frequency; PTEN, Phosphatase and Tensin homolog; TP53, tumor protein 53; DNA, deoxyribonucleic acid; PCR, polymerase chain reaction; NGS, next generation sequencing.

DCE-MRI was performed in 8 ovarian cancer cases to explore vascular changes. Prior studies have shown malignant tumors typically exhibit more rapid and intense enhancement on DCE-MRI compared to benign lesions.²¹ We examined the correlation between K^trans changes and treatment response. Most patients demonstrated stable disease, limiting conclusions about DCE-MRI’s predictive value. Despite the small sample size, the results (Figure 2) suggest a potential trend: greater reduction in K^trans appeared associated with reductions in target lesion size (RECISTv1.0), while smaller K^trans changes trended with lesion growth. However, a single outlier with substantial tumor growth despite modest K^trans reduction weakens the trend. In this DCE-MRI subgroup, the positive mean ΔK^trans of 2.40 ± 5.40 and moderate tumor size change (8.25 ± 7.23 cm), suggests that the tumor environment is relatively stable. While RECIST remains the standard for assessing radiographic response, incorporating K^trans measurements offers additional insight into vascular changes and perfusion dynamics not readily captured by size criteria alone. These early microcirculatory changes could enhance sensitivity to treatment effects and potentially inform earlier modifications to therapy.

Figure 2.

Results of DCE-MRI for each patient subgroup.

Figure 1.

(A) Waterfall plot depicting best overall RECISTv1.0 response; (B) Swimmer plot showing duration of therapy; (C) Oncoprint diagram.

Resource and logistical constraints limited full DCE-MRI analysis and longitudinal analysis, reducing the number of evaluable cases. Additionally, patients enrolled had extensive metastatic disease and were heavily pre-treated. While the findings of this study provide valuable insights into the therapeutic potential of DCE-MRI, it is crucial to interpret the results considering these limitations.

This study has been completed, and no further testing of the combination is planned given the modest clinical activity observed.

Author contributions

Sarina A. Piha-Paul (Supervision, Conceptualization, Funding acquisition, Investigation, Methodology, Writing—original draft, Writing—review & editing) Chieh Tseng, Hung Le, Lei Kang and Emily Thompson (Formal analysis, Writing—original draft, Writing—review & editing), R. Jason Stafford (Supervision, Writing-review & editing), Siqing Fu, Apostolia Tsimberidou, George Blumenschein, Jordi Rodon Ahnert, John M. Slopis, David Hong, Aung Naing, Funda Meric-Bernstam, Chaan S. Ng, Shannon Westin, and Anil K. Sood (Investigation, writing-review & editing)

Funding

The study is supported, in part by NIH/NCI CA281701 (A.K.S), NIH/NCI P30CA016672-Core Grant (CCSG Shared Resources), American Cancer Society (A.K.S), NIH Clinical and Translational Science Award Grants UM1TR0045906 and 1UL1 TR003167.

Conflicts of interest

S.A. Piha-Paul has received research and grant funding through MD Anderson from ABM Therapeutics, Inc.; Alkermes; Aminex Therapeutics; Axcynsis Therapeutics Pte. Ltd; BioMarin Pharmaceutical, Inc; Boehringer Ingelheim; Cullinan Oncology, Inc; Chugai Pharmaceutical Co., Ltd; Cyclacel Pharmaceuticals; Day One Biopharmaceuticals Inc; Daiichi Sankyo; ENB Therapeutics; Epigenetix Inc.; Genmab US, Inc.; Gilead Sciences, Inc.; Hengrui Pharmaceuticals, Co., Ltd; Immunity Bio, Inc.; Immunome, Inc.; Immunomedics, Inc.; Incyte Corp.; Innovent Biologics Co., Ltd; iTeos Belgium SA; Jazz Pharmaceuticals; Johnson & Johnson; Loxo Oncology, Inc.; Merck Sharp and Dohme Corp.; Mitsubishi Tanabe Pharma America (MTPA) Inc.; Nectin Therapeutics, Ltd; Nested Therapeutics, Inc.; NRG Oncology; Nurix; OncoNano Medicine, Inc; Pyxis Oncology; Pieris Pharmaceuticals, Inc.; Pfizer; Phanes Therapeutics; Puma Biotechnology, Inc.; Purinomia Biotech, Inc.; Replimune; Roche/Blueprint; Solve Therapeutics; Strand Therapeutics, Inc.; Tallac Therapeutics, Inc.; Toragen Therapeutics, Inc.; TransThera Bio; Xencor, Inc; NCI/NIH: P30CA016672—Core Grant (CCSG Shared Resources); CPRIT Grant: Cancer Prevention Research Institute of Texas (CPRIT) Precision Oncology Decision Support Core (RP150535) and Clinical and Translational Science Award Grant (CTSA): 1UM1TR0045906. She has also worked as a consultant for Mitsubishi Tanabe Pharma America (MTPA) Inc. D. Hong reports research(Inst)/grant funding (Inst): AbbVie, Adaptimmune, Adlai-Nortye, Amgen, AstraZeneca, Bayer, Biomea, Bristol Myers Squibb, Daiichi-Sankyo, Deciphera, Eisai, Eli Lilly, Endeavor, Erasca, F. Hoffmann-LaRoche, Fate Therapeutics, Genentech, Genmab, Immunogenesis, Infinity, Kyowa Kirin, Merck, Mirati, Navier, NCI-CTEP, Novartis, Numab, Pfizer, Pyramid Bio, Revolution Medicine, SeaGen, STCube, Takeda, TCR2, Turning Point Therapeutics, VM Oncology; travel, accommodations, expenses: AACR, ASCO, CLCC, Bayer, Genmab, SITC, Telperian; consulting, speaker, or advisory role: 28Bio, Abbvie, Acuta, Adaptimmune, Alkermes, Alpha Insights, Amgen, Affini-T, Astellas, Aumbiosciences, Axiom, Baxter, Bayer, Boxer Capital, BridgeBio, CARSgen, CLCC, COG, COR2ed, Cowen, Ecor1, Erasca, Fate Therapeutics, F.Hoffmann-La Roche, Genentech, Gennao Bio, Gilead, GLG, Group H, Guidepoint, HCW Precision Oncology, Immunogenesis, InduPro, Janssen, Liberium, MedaCorp, Medscape, Numab, Oncologia Brasil, ORI Capital, Pfizer, Pharma Intelligence, POET Congress, Prime Oncology, Projects in Knowledge, Quanta, RAIN, Ridgeline, SeaGen, Stanford, STCube, Takeda, Tavistock, Trieza Therapeutics, Turning Point Therapeutics, WebMD, YingLing Pharma, Ziopharm; other ownership interests: Molecular Match (advisor), OncoResponse (founder, advisor), Telperian (founder, advisor). J. Rodon Ahnert reports personal fees from Ellipses Pharma, IONCTURA, Clarion Healthcare, Debiopharm, Monte Rosa Therapeutics, Cullgen, Macrogenics, Oncology One, Envision Pharma, Columbus Venture Partners, Sardona Therapeutics, Avoro Capital Advisors, Vall d‘Hebron Institute of Oncology/Ministero De Empleo Y Seguridad Social, Chinese University of Hong Kong, Boxer Capital, LLC, Tang Advisors, LLC, Incyte, Alnylam Pharmaceuticals; grants and personal fees from Aadi Bioscience, Pfizer, Merus; grants from Blueprint Medicines, Black Diamond Therapeutics, Merck Sharp & Dohme, Hummingbird, Yingli, Vall d‘Hebron Institute of Oncology/Cancer Core Europe, Novartis, Spectrum Pharmaceuticals, Symphogen, BioAtla, GenMab, CytomX, Kelun-Biotech, Takeda-Millennium, GlaxoSmithKline, Taiho, Roche Pharmaceuticals, Bicycle Therapeutics, Curis, Bayer, Nuvation, ForeBio, BioMed Valley Discoveries, Loxo Oncology, Hutchinson Medipharma, Cellestia, Deciphera, Ideaya, Amgen, Tango Therapeutics, Mirati, Linnaeus Therapeutics; other from European Society for Medical Oncology and Vall d‘Hebron Institute of Oncology/Ministero De Empleo Y Seguridad Social outside the submitted work. S. Fu reports research funding from Abbisko (Inst), Anaeropharma (Inst), Arrien Pharmaceuticals (Inst), AstraZeneca (Inst), BeiGene (Inst), BioAtla (Inst), Boehringer Ingelheim (Inst), Hookipa Pharma (Inst), Huya Bioscience International (Inst), IMV (Inst), Innovent Biologics (Inst), Lilly (Inst), Lyvgen Biopharma (Inst), MacroGenics (Inst), Medivir (Inst), NCCN (Inst), NCCN (Inst), Nerviano Medical Sciences (Inst), NeuPharma, Inc. (Inst), NIH/NCI (Inst), Novartis (Inst), NovoCure (Inst), OncoMed (Inst), Parexel International, LLC (Inst), Sellas Life Sciences (Inst), Soricimed (Inst), Taiho Oncology (Inst), Tolero Pharmaceuticals (Inst) and Turnstone Bio (Inst). A. Tsimberidou served as consulting or advisory role for Vincerx, BrYet and Diaccurate, NEX-I, Macrogenics, Avstera. A. Naing Niang reports grants from NCI, EMD Serono, MedImmune, Healios Onc. Nutrition, Atterocor/Millendo, Amplimmune, ARMO BioSciences, Karyopharm Therapeutics, Incyte, Novartis, Regeneron, Merck, Bristol Myers Squibb, Pfizer, CytomX Therapeutics, Neon Therapeutics, Calithera Biosciences, TopAlliance Biosciences, Eli Lilly, Kymab, PsiOxus, Arcus Biosciences, NeoImmuneTech, Immune-Onc Therapeutics, Surface Oncology, Monopteros Therapeutics, BioNTech SE, Seven & Eight Biopharma and SOTIO Biotech AG and GV20 Therapeutic; consulting fees from CTI, Deka Biosciences, Janssen Biotech, Mural Oncology, NGM Bio, PsiOxus Therapeutics, Immune-Onc Therapeutics, STCube Pharmaceuticals, OncoSec KEYNOTE-695, Genome & Company, CytomX Therapeutics, Nouscom, Merck Sharp & Dohme Corp, Servier, Lynx Health, AbbVie. Travel expenses from ARMO BioSciences, NeoImmuneTech, NGM Biopharmaceticals; Honoraria for speaking engagements from AKH Inc, The Lynx Group, Society for Immunotherapy of Cancer (SITC), Korean Society of Medical Oncology (KSMO), Scripps Cancer Care Symposium, ASCO Direct Oncology Highlights, European Society for Medical Oncology (ESMO), CME Outfitters outside the submitted work; and spouse research funding: The Texas Medical Center Digestive Diseases Center, Jeffery Modell Foundation, Immune Deficiency Foundation, Baxalta US Inc, Chao Physician-Scientist Foundation; consultant/advisory board: Takeda, Pharming Healthcare Inc and Horizon Therapeutics USA, Inc.; ad hoc consultancy speaker: Alfaisal University. F. Meric-Bernstam reports research funding/grant support for clinical trials from Aileron Therapeutics, AstraZeneca, Bayer Healthcare Pharmaceutical, Calithera Biosciences, Curis, CytomX Therapeutics, Daiichi Sankyo, Debiopharm International, eFFECTOR Therapeutics, Genentech, Guardant Health, Klus Pharma, Takeda Pharmaceutical, Novartis, Puma Biotechnology, Taiho Pharmaceutical. F. Meric-Bernstam served on advisory committee for Black Diamond, Biovica, Eisai, FogPharma, Immunomedics, Inflection Biosciences, Karyopharm Therapeutics, Loxo Oncology, Mersana Therapeutics, OnCusp Therapeutics, Puma Biotechnology Inc., Seattle Genetics, Sanofi, Silverback Therapeutics, Spectrum Pharmaceuticals, Zentalis, personal fees for consulting/travel related from AbbVie, Aduro BioTech, Alkermes, AstraZeneca, Daiichi Sankyo, DebioPharm, Ecor1 Capital, eFFECTOR Therapeutics, F. Hoffman-La Roche, GT Apeiron, Genentech, Harbinger Health, IBM Watson, Infinity Pharmaceuticals, Jackson Laboratory, Kolon Life Science, Lengo Therapeutics, Menarini Group, OrigiMed, PACT Pharma, Parexel International, Pfizer, Protai Bio, Samsung Bioepis, Seattle Genetics, Tallac Therapeutics, Tyra Biosciences, Xencor, Zymeworks, EuropeanOrganization for Research and Treatment of Cancer (EORTC), European Society for Medical Oncology (ESMO), personal fees for honoraria from Chugai Biopharmaceuticals. A.K. Sood has received consulting fees from (Immunogen, Onxeo); IDMS (Advenchen); DSMB (Mural Oncology). All remaining authors have declared no conflicts of interest.

Data Availability

The de-identified participant data and dataset generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.