If It’s a Solid Tumor Target, It May Be a Hematologic Cancer Target: Bridging the Great Divide

SUMMARY

Tumor-agnostic Food and Drug Administration approvals are transforming oncology. They include larotrectinib/entrectinib/repotrectinib, NTRK fusions; selpercatinib, RET fusions; dabrafenib/trametinib, BRAF^V600E mutations; pembrolizumab/dostarlimab, microsatellite instability; pembrolizumab, high tumor mutational burden; and trastuzumab deruxtecan for HER2 3+ expression (all solid cancers). Pemigatinib is approved for FGFR1-rearranged myeloid/lymphoid neoplasms. The genomically-driven tissue-agnostic approach has a strong biological rationale (cancer is a disease of the genome), yields remarkably high response rates, and provides drug access to patients with an unmet need (rare/ultra-rare malignancies). Despite the solid tumor focus, both solid and hematologic cancers can harbor identical driver molecular abnormalities and respond to cognate therapies. For example, BRAF^V600E and IDH1/2 mutations, ALK, FGFR and NTRK fusions, PD-L1 amplification, and CD70 antigen, are druggable in both solid and blood malignancies by gene-/immune-targeted therapies/chimeric antigen receptor T cells. Future biomarker-based tissue-agnostic basket studies/approvals should bridge the great divide and include both solid and hematologic cancers.

eTOC Blurb

Targeted, immune, and cellular therapies have great therapeutic potential for patients with cancer. Molecular biomarkers may enhance the use of these therapies in a tumor-agnostic fashion across both solid and hematologic malignancies.

INTRODUCTION

Tumor-agnostic/tissue-agnostic Food and Drug Administration (FDA) approvals, meaning that a therapeutic is approved regardless of tissue of origin providing that the cancer bears the requisite molecular biomarker, are rapidly transforming the cancer treatment landscape. Importantly, powerful genomic sequencing techniques such as next generation sequencing (NGS) demonstrate that cancer is a disease of the genome.^1–3 Hence, over the last decade, cancer treatment has undergone a radical makeover, moving from an organ-of-origin (tumor-specific)/molecular-agnostic approach to a tumor-specific, molecular-specific paradigm and more recently to a molecular-specific/tumor-agnostic therapeutic paradigm.⁴ It remains to be seen how often molecularly targeted therapies will be effective across tumor types. Basket trials, which are a type of clinical trial encompassing a wide range of tumor types that have a common biomarker, are needed to speed the process of answering this question.

Molecular-specific/tumor-agnostic therapies emerged largely from certain clinical observations and needs⁵: (i) finding driver molecular alterations across tumor types; (ii) discovering that these molecular alterations are druggable across tumor types; (iii) learning that many molecular alterations are rare, making it virtually impossible to perform clinical trials in every tissue-of-origin tumor type that might potentially harbor the alteration (which might number several hundred types of tumors, some of which are rare or ultra-rare even before the molecular subset is isolated): (iv) establishing the scientific principle that targeted drugs acted on the molecular target, not on a specific tissue; and (v) recognizing that targeted drugs had anti-cancer activity when given to patients whose cancers harbored the molecular target, often (but not always) regardless of tissue of origin. Therefore, the development of gene- and immune-targeted drugs coupled with the molecular re-classification of cancer has yielded a new treatment era.^6,7

Until now, tumor-agnostic biomarker-based approvals have focused for the most part of solid cancers. Yet, many of the molecular targets are also seen in hematologic malignancies, which are also responsive to the matched targeted therapeutic (Figure 1). Herein, we provide an overview of the field of tissue-agnostic targets, in particular as it pertains to both solid and hematologic cancers, with the basic hypothesis being that if it’s a solid tumor target, it may also be a hematologic target (and vice versa) – tissue is not the issue.

Figure 1. Bridging the Great Divide.

Solid tumors and hematologic malignancies are studied in separate trials and no tissue-agnostic approvals span both types of cancer. Yet, many molecular alterations, including but not limited to ALK, FGFR, and NTRK fusions, BRAF^V600E and IDH1/2 mutations, and PD-L1 amplification as well as cell-surface antigens such as CD70 are seen and are actionable in both solid and hematologic cancer. Created with BioRender.com.

TISSUE-AGNOSTIC FDA APPROVALS: BREADTH AND DEPTH

There are now multiple tissue-agnostic Food and Drug Administration (FDA) approvals, particularly for solid malignancies. The list of approvals includes antibodies that act as immune checkpoint inhibitors (ICIs): pembrolizumab (anti-PD-1) for high tumor mutation burden (TMB) (≥10 mutations/mb) and microsatellite unstable (MSI-H)/deficient mismatch repair (dMMR) solid cancers as well as dostarlimab (anti-PD-1) for MSI-H/dMMR solid tumors. In addition, drugs (small molecule inhibitors) targeting NTRK fusions (larotrectinib, entrectinib, and repotrectinib), RET fusions (selpercatinib), and BRAF^V600E mutations (dabrafenib [BRAF inhibitor] given together with trametinib [MEK inhibitor]), as well as deruxtecan trastuzumab (antibody-drug conjugate) for HER2 3+ (by immunohistochemistry) have also attained FDA solid tumor-agnostic approval.^5,8–14

Regulatory authorization has often been based on small single-arm studies or compiling the data from several studies, generally with a total of ~30-~150 patients. The studies show high objective response rates (ORRs), in the range of ~30% to ~45% for the immunotherapy tissue-agnostic approvals and ~40% to ~85% for the gene-targeted tissue-agnostic approvals.¹⁵ Moreover, the duration of response is often long, even in patients with widely metastatic disease that is refractory to numerous traditional therapies. As an example, the NTRK-fusion solid tumor-agnostic approval for larotrectinib was based on a collation of three trials that included 12 cancer types amongst 55 patients (12 of whom were children); the ORR was 75%, including 22% complete responses and 53% partial responses. Response duration was ≥6 months for 73% and ≥12 months for 39% of patients.¹⁶ Similarly, for pembrolizumab in MSI-H/dMMR solid tumors, approval was based on data from 149 patients with MSI-H/dMMR cancers enrolled across five multi-cohort, multi-center, single-arm clinical trials. ORR was 39.6% and responses lasted ≥6 months for 78% of those who responded to pembrolizumab. ORR was similar irrespective of whether patients were diagnosed with colorectal cancer (36%) or a different cancer type (46% across the 14 other cancer types).⁸ For dostarlimab’s MSI-H solid tumor-agnostic approval, ORR was 45.4%, with a 15.6% complete response rate; 85.9% of patients had a duration of response ≥12 months and 54.7% of patients had durations of response ≥24 months.¹⁷ For TMB-High (≥10 mutations/mb), the ORR in 102 patients with a variety of cancers was 29% after pembrolizumab treatment; remarkably, 57% of patients had response durations ≥12 months and 50% of patients had response durations ≥24 months. Other studies have confirmed a relationship between TMB and response to ICIs, with some reports suggesting that the relationship is linear.¹⁸ A higher TMB cut off may yield higher response rates. For instance, in 42 patients with diverse cancer types treated with the anti-PD-L1 agent atezolizumab and with TMB ≥16 mut/Mb, ORR was 38.1% versus 2.1% for 48 patients with TMB ≥10 and <16 mut/Mb.¹⁹ Importantly, a subset of patients with widely metastatic, treatment-refractory high TMB and/or MSI-H tumors treated with ICIs can achieve durable complete remissions — cures.²⁰

Tissue-agnostic approvals have wide breadth. For instance, in most cases, they include both children and adults, though the lower age limit may vary. Indeed, for solid tumor-agnostic approvals, only those for dostarlimab (MSI-H solid cancers), selpercatinib (RET fusion-bearing solid cancers), and deruxtecan trastuzumab (HER2 3+ immunohistochemistry) are specific only to adults.¹⁵ Moreover, the FDA has shown flexibility in other ways. For instance, for the approval of the dabrafenib/trametinib combination for BRAF^V600E-mutated solid cancers, colorectal cancer was excluded because BRAF^V600E-bearing colorectal cancers had lower response rates (~12%) to dabrafenib and trametinib as compared to other solid tumor types (~41%).^21,22 The exclusion of colorectal cancer permitted the solid tumor-agnostic BRAF^V600E approval to proceed for other tissue types.

All tissue-agnostic approvals require a cognate molecular biomarker. Some tissue-agnostic approvals focus on cancers harboring rare/ultra-rare biomarkers (<1% of malignancies), compelling NGS interrogation of hundreds of cancers to find a single one bearing the molecular anomaly. As examples, NTRK fusions are observed in 0.31% of adult tumors and in 0.34% of pediatric tumors, and RET fusions occur in about 0.6% of cancers.^23–25 Notably, although many molecular abnormalities are discerned in both solid and hematologic malignancies, most tissue-agnostic approvals focus on solid cancers.

WHAT ABOUT HEMATOLOGIC MALIGNANCIES?

The field of hematologic malignancies led the way for many aspects of the molecular revolution in cancer therapy. For instance, in the 1990s and earlier, patients with chronic myelogenous leukemia (CML) inevitably succumbed within about 3 to 4 years, but now have a near-normal life expectancy.²⁶ This dramatic improvement in outcome is due to several factors: the discovery of the aberrant Bcr-Abl kinase (produced by the BCR-ABL gene fusion) that drives CML; finding that drugs, starting with imatinib, are potent inhibitors of the Bcr-Abl kinase; and moving Bcr-Abl-targeted treatment to early stages of the disease.^27–30 Even so, tumor-agnostic gene- or immune-targeted therapies have lagged in the hematologic malignancies.

Gene-targeted therapies:

Despite multiple tissue-agnostic FDA approvals for various solid tumors, there is only one tumor-agnostic approval for hematologic malignancies — pemigatinib for myeloid/lymphoid neoplasms (MLNs) with FGFR1 rearrangement—an ultra-rare subgroup. In a study from the Mayo Clinic involving over 24,000 sequential cases of leukemia analyzed, only ~0.02% showed an FGFR1 rearrangement.³¹ When the FGFR inhibitor pemigatinib was given to patients with myeloid/lymphoid neoplasms and FGFR1 rearrangement in the study known as FIGHT-203, the complete cytogenetic response rate was ~79%. This remarkable result, in only 28 patients, led to FDA approval.^32,33 Of interest, pemigatinib also has a biomarker-based solid cancer approval – for FGFR2 fusion-bearing cholangiocarcinoma.³⁴ FGFR1 fusions are also found in in 0.18% of AACR GENIE cases, most commonly in breast, colon, lung, and prostate cancer.³⁵

Importantly, there has yet to be a tumor-agnostic approval for solid tumors that includes a hematologic malignancy, despite the fact that activity is seen in hematologic malignancies (Table 1^{27,31,32,36–95}). For example, BRAF^V600E alterations have been documented in patients with hairy cell leukemia (with virtually 100% of patients showing the alteration), Erdheim-Chester disease (ECD), and in a subset of myelomas, as well as in other hematologic malignancies, albeit rarely.^38,42,47,49 Although not FDA approved, the use of the BRAF inhibitor vemurafenib in patients with relapsed or refractory hairy cell leukemia in two phase 2, multicenter studies with 26 and 24 patients, respectively, showed remarkable ORRs of 96% and 100%.⁴⁰ Similarly, not FDA approved, an open-label, phase II trial of dabrafenib with trametinib in 55 patients with refractory hairy cell leukemia showed an ORR of ~89%.⁴¹ The FDA did approve the use of the BRAF inhibitor vemurafenib in patients with the non-Langerhans histiocytosis ECD, based on an open-label, multicenter, single-arm, multiple cohort clinical trial in 22 patients ≥ 6 years old with an ORR of ~55%.⁴³ Similarly, in ECD, the MEK inhibitor cobimetinib was approved based on a single-center, single-arm trial of 26 patients with an ORR of ~77%.⁴⁸ Finally, the MEK inhibitor trametinib has been listed in the National Comprehensive Cancer Network (NCCN) guidelines for ECD and shows a 71% ORR with a median time-to-treatment failure of 37 months.⁹⁶

Table 1.

Examples of established and potential targets in hematologic malignancies with and without counterparts in solid cancers

Disease	Molecular alterations	Frequency	Responsiveness to targeted drugs	FDA Approvals	Comments (and solid cancer counterparts)
Hairy cell leukemia	BRAF V600E	~100%38	BRAF inhibitor vemurafenib (~96% ORR)39,40	There are no BRAF targeted therapies approved by the FDA for hairy cell leukemia	BRAFV600E has a tissue-agnostic FDA approval in solid cancers other than colorectal cancer (dabrafenib combined with trametinib)
			BRAF inhibitor dabrafenib + MEK inhibitor trametinib (~89% ORR)41
Erdheim-Chester disease (non-Langerhans histiocytosis)	BRAF V600E	~54%42	BRAF inhibitor vemurafenib (~55–60% ORR)43	FDA approved44	These patients often require significantly reduced doses to avoid toxicity46
			Cobimetinib (89% ORR in histiocytic neoplasms regardless of BRAF status); these neoplasms typically have BRAF pathway alterations or MEK pathway alterations)45–48,96	Treatment of adult patients with histiocytic neoplasms, which include ECD, Rosai-Dorfman disease, and Langerhans cell histiocytosis93
Myeloma	BRAF V600E	2.8%49	Vemurafenib (33% ORR)50	No approval for BRAF directed therapy in myeloma
			BRAF inhibitor encorafenib + MEK inhibitor cobimetinib (83.3% ORR)51
	FGFR1–3 alterations	40%52	NCI-MATCH included myeloma but accrued no myeloma patients.53	No approval for FGFR directed therapy in myeloma
Chronic lymphocytic leukemia	BRAF alterations	2%54	No trials	No approval for therapy in CLL targeting these alterations
	KITLG, KIT, PTPN11, GNB1, KRAS NRAS	2.6%54
	MAPK2K1, MAPK2K2, and MAPK1	1.1%54
Myeloid/Lymph oid Neoplasms (MLNs)	FGFR1 rearrangement	0.000164%31	Pemigatinib (>70% complete cytogenetic response rate)37	FDA approval (hematologic tissue -agnostic) for adults with relapsed or refractory MLNs with FGFR1 rearrangement32 This is the only hematologic tissue-agnostic approval	Pemigatinib is also approved for cholangiocarcinoma with an FGFR2 fusion or rearrangement36
CML	BCR-ABL fusion (P210Bcr-Abl)	~100%27,55	Imatinib (65%−83% complete cytogenetic response rate in newly diagnosed chronic phase disease56–59	Newly diagnosed adult patients with Ph+ CML in chronic phase. Patients with Ph+ CML in blast crisis, accelerated phase, or in chronic phase after failure of interferon-alpha therapy	A patient with glioblastoma multiforme and a BCR-ABL fusion responded to imatinib104 Note: P190Bcr-Abl also found in a subset of patients with AML71
			Nilotinib (80% complete cytogenetic response rate in newly diagnosed chronic phase CML59,60	Treatment of adults with chronic-phase CML who are resistant or intolerant to prior therapies Including imatinib
			Dasatinib (77% complete cytogenetic response rate in newly diagnosed chronic phase CML)58,61	Treatment of adults with CML who are resistant or intolerant to prior therapies Including imatinib
			Ponatinib (46% complete cytogenetic response in chronic-phase CML with resistance to or unacceptable side effects from dasatinib or nilotinib or who had the Bcr-Abl T315I mutation.)62	Chronic-phase CML with resistance or intolerance of at least 2 prior kinase inhibitors, with appropriate monitoring for arterial occlusive events
			Bosutinib (46–64% major cytogenetic response in chronic-phase CML resistant or intolerant to imatinib63 ~77% complete cytogenetic response rate in newly diagnosed chronic phase CML64	Chronic- phase CML patients with resistance or intolerance to imatinib
			Asciminib (40.8% complete cytogenetic response in chronic phase CML resistant/intolerant to ≥2 tyrosine kinase inhibitors)65	Chronic- phase CML, previously treated with tyrosine kinase inhibitors, and for adults (chronic phase) with the T315I mutation
ALL	BCR-ABL fusion (P190Bcr-Abl)	∼3%−5% of pediatric ALL and 25% of adult ALL66,67	Ponatinib (47% major cytogenetic response in Ph-positive ALL with resistance to or unacceptable side effects to dasatinib or nilotinib or with the T315I mutation)62	Ph+ ALL for whom no other tyrosine kinase inhibitor therapy Indicated or T315I mutation positive
			Imatinib plus chemotherapy (89% complete remission in newly diagnosed with Ph+ ALL)68	Children newly diagnosed with Ph+ ALL
			Nilotinib + chemotherapy (53% molecular complete remission in newly diagnosed with Ph+ ALL)69	Newly diagnosed with Ph+ ALL
			Dasatinib + chemotherapy (88% complete remission in newly diagnosed Ph+ ALL)70	In combination with chemotherapy for the treatment of children ≥1 year old with newly diagnosed Ph+ ALL
AML	BCR-ABL fusion (P190Bcr-Abl)	0.5%−3%71,72	Ponatinib + venetoclax (43% ORR in Ph+ AML)73	No approval for Bcr-Abl-bearing AML
APL	PML-RARA fusion	~100%74	All-trans retinoic acid + arsenic trioxide (100% complete remission)75 All-trans retinoic acid (72% complete remission)77	FDA approved, first-line treatment of adults with newly diagnosed, low-risk acute promyelocytic leukemia (APL) and t(15;17) or PML-RARA gene expression
Multiple hematologic malignancies	NTRK fusions	0.1%78,79	Anecdotal refractory AML and ALL with -NTRK2 fusion responded to NTRK inhibitor	No approval in hematologic malignancies for NTRK inhibitors	NTRK fusions have two tissue-agnostic FDA approvals for solid tumors (drugs, larotrectinib and entrectinib)
AML	IDH1 mutations	6–16%80	IDH1-mutated AML has response rate of 32.8% to IDH1 inhibitor ivosidenib81	Ivosidenib is FDA approved for IDH1-mutated AML81	Ivosidenib is FDA approved for IDH1-mutated cholangiocarcinoma82
AML	IDH2 mutations	8–19%	IDH2-mutated AML has a response rate of 23% to the IDH2 inhibitor enasidenib83	Enasidenib is FDA approved for IDH1-mutated AML	Vorasidenib, an investigational IDH1/2 inhibitor significantly improved progression-free survival in grade 2 IDH1/2-mutated gliomas.84,85
ALCL	ALK fusions	~50–80%	ALCL has a response rate of 88% to the ALK inhibitor crizotinib.86	Crizotinib is FDA approved for ALCL86	Multiple ALK inhibitors (alectinib, brigatinib, ceritinib, crizotinib, and lorlatinib) are FDA approved for ALK-rearranged NSCLC with response rates in the 50–70% range.85
Myeloid lymphoid neoplasms	FGFR1 fusions	0.02%	79% response rate to FGFR inhibitor pemigatinib	Pemigatinib is FDA approved for FGFR1 rearranged myeloid lymphoid malignancies	Pemigatinib is FDA approved for FGFR2-rearranged cholangiocarcinomas. FGFR1 fusions are found in in 0.18% of AACR GENIE cases, being most frequent in breast, colon, lung and prostate cancer.35
Hodgkin lymphoma	PD-L1 (CD274) amplification	Almost 100% of cases of classic Hodgkin lymphoma.87	Hodgkin lymphoma (treatment-refractory) has a response rate of 65–87% to ICIs.88,89	Anti-PD-1 agents pembrolizumab and nivolumab are approved for Hodgkin lymphoma90,91	ORR for patients with solid tumors that harbored PD-L1 amplification and were treated with ICIs was 66.7% (6 of 9 patients).92
Lymphomas and AML	CD70	CD70 is expressed by Hodgkin and non-Hodgkin lymphoma, Waldenström macroglobulinemia, multiple myeloma, and by human T-lymphotropic virus type 1– and EBV-associated malignancies.94	CAR T-cell clinical trials are underway		CD70 is aberrantly expressed in solid tumors, including renal cell cancer, glioblastomas, osteosarcoma, thymic carcinoma, nasopharyngeal carcinoma, ovarian, lung, colon and pancreatic cancer, and melanoma with highest expression in renal cell carcinoma (79.5%).95 CAR T-cell trials in both hematologic and solid cancers are ongoing (NCT04438083; NCT05947487; NCT04502446)

Abbreviations: ALCL, anaplastic large cell lymphoma; ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; APL, acute promyelocytic leukemia; CLL, chronic lymphocytic leukemia; CML, chronic myeloid leukemia; ECD, Erdheim-Chester Disease; FDA, Food and Drug Administration; ICI= immune checkpoint inhibitors; MLNs, myeloid/lymphoid neoplasms; MRD, minimal residual disease; NSCLC= non-small cell lung cancer; ORR, objective response rate; Ph+, Philadelphia chromosome positive; PML-RARA, promyelocytic leukemia-retinoic acid receptor alpha

In BRAF^V600E-positive multiple myeloma, nine patients received dabrafenib with trametinib and an ORR of 33% was observed.⁵⁰ Additionally, the combination of the BRAF inhibitor encorafenib and the MEK inhibitor cobimetinib in 12 heavily pretreated (median 5 lines, range 2–14) patients with BRAF V600E-positive multiple myeloma resulted in an ORR of 83.3%, albeit with a relatively short progression-free survival of 5.6 months; overall survival was 55% at 24 months.⁵¹

Taken together, these observations indicate that BRAF alterations have been characterized in various malignant hematologic conditions. Indeed, the malignancy with both the highest rate of BRAF^V600E and highest response rate to BRAF inhibitors is a hematologic cancer – hairy cell leukemia—with virtually 100% of patients having the BRAF^V600E and nearly all responding to BRAF pathway inhibitors.^38,40 Therefore BRAF/MEK inhibitors warrant study in a broader range of hematologic cancers, with the question as to whether a basket trial could be the basis of expanding the tumor-agnostic approval to hematologic malignancies.⁴⁵

A similar question could be asked for NTRK fusions. In an evaluation of >7,000 patients with hematologic malignancies, NTRK fusions were discerned in a small subset (0.1%) of patients with lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), histiocytosis, multiple myeloma, and dendritic cell neoplasms. NTRK fusions conferred responsiveness to NTRK inhibition in vitro and in vivo in a patient-derived xenograft and in anecdotal AML and ALL patients with NTRK-fusions treated with larotrectinib.^78,79

Similarly, IDH1 mutations have been found in solid tumors and in hematologic malignancies, and the FDA has approved the IDH1 inhibitor ivosidenib in IDH1-mutated AML and in IDH1-mutated cholangiocarcinoma^97,98 In IDH1-mutated AML, approval was based on a single-arm, multicenter clinical trial that included 174 adult patients with relapsed or refractory AML with an IDH1 mutation. The rate of complete remission plus complete remission with partial hematologic recovery was 32.8%. Ivosidenib is also FDA approved for IDH1-mutated cholangiocarcinoma based on a randomized trial that showed significant improvement in progression-free survival.⁸² IDH2 mutations are also found in both solid tumors and hematologic malignancies. Vorasidenib, an investigational IDH1/2 inhibitor significantly improved progression-free survival and delayed the time to the next intervention in grade 2 IDH1/2-mutated gliomas.⁸⁴ Enasidenib, an IDH2 inhibitor is FDA approved for IDH2-mutated AML based on a response rate (complete response and complete response with partial hematologic recover) of 23%.⁸³

Activating ALK alterations (amplifications, mutations, and fusions/rearrangements) are detectable in numerous cancers including, but not limited to, non-small lung cancer (NSCLC), anaplastic large-cell lymphoma (ALCL) (an uncommon, aggressive CD30-positive T-cell lymphoma comprising 0.5% of adult lymphomas and ~10% of non-Hodgkin lymphoma pediatric cases), inflammatory myofibroblastic tumors (IMT) (rare intermediate-grade neoplasms, generally found in children), neuroblastomas, and inflammatory breast cancers. ALK fusions are particularly sensitive to ALK inhibitors.^99,100 ALK fusions/rearrangements are rare, being detected in ~0.5–0.8% of cancers.^35,101 Among patients with NSCLC, the frequency of ALK fusions/rearrangements is >3%; in contrast, the frequency in non-NSCLC tumors is about ~0.2%. Besides NSCLC, IMT (~50% have ALK fusions/rearrangements) and ALCLs (~50–80% having ALK fusions/rearrangements) are the neoplasms most often harboring ALK fusions.⁸⁵ The FDA has approved the ALK inhibitor crizotinib for the treatment of the hematologic malignancy ALCL (pediatric patients 1 year of age and older and young adults) based on an ORR of 88% in 26 patients.^86,102,103 Multiple ALK inhibitors (alectinib, brigatinib, ceritinib, crizotinib, and lorlatinib) are also approved for the treatment of ALK-rearranged NSCLC with response rates in the 50–70% range.⁸⁵ Hence, ALK fusion/rearrangement is another target that is pharmacologically tractable in both solid and hematologic malignancies.

Finally, the BCR-ABL fusion is known to occur almost exclusively in leukemias.³⁰ However, recently, a glioblastoma bearing a BCR-ABL fusion was reported and shown to respond to the Bcr-Abl kinase inhibitor imatinib.¹⁰⁴ This case demonstrates that in the rare instance when a genomic alteration that appears specific to a certain tumor type is seen in a different tumor type, it may still be pharmacologically tractable.

Immune checkpoint inhibitors:

Biomarkers associated with response to ICIs include high TMB (regardless of MSI-H status), MSI-H (which often leads to high TMB)¹⁰⁵, and PD-L1 immunohistochemistry (IHC) positivity^106,107 Large-scale studies show that ~1.5% of cancers have MSI-H and about 7% have TMB≥20 mutations/mb and ~13% have TMB≥10 mutations/mb (the latter being the cut-off for the tumor-agnostic approval of pembrolizumab for solid cancers).^105,108,109 Across tumor types, outcome is significantly better for MSI-H-bearing tumors that received immunotherapy compared to non-immunotherapy.¹¹⁰ Most of these studies have focused on solid cancers.

In regard to blood-based cancers, as an example, in one study, 390 patients with various hematologic malignancies interrogated by NGS were identified. Forty-eight of the 390 samples (12%) had a high TMB (≥10 mutations/mb). Twenty-five of 45 (56%) patients with diffuse large B-cell lymphoma had a high TMB. The TMB was low in all myeloid malignancies tested. None of the 302 samples tested were MSI-H. PD-L1 IHC was performed on 86 samples. Eleven (13%) had high expression (≥ 50%), 26 (30%) had low expression (1–49%), and 49 (57%) had no expression of PD-L1 on the tumor cells. The majority of samples with PD-L1 expression were mature lymphomas (81%). TMB and PD-L1 score had a significant linear relationship (R = 0.22, p= 0.04, 95% CI 0.01 – 0.41).¹¹¹ In one report, up to 50% of therapy-related acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) cases exhibit an MSI-H phenotype (as compared with <5% of de novo AML/MDS cases).^112,113 The prevalence of MSI-H is highly variable across other blood cancer types; it is very low (< 1%) in non-Hodgkin lymphomas, except immunodeficiency-related lymphomas.¹¹⁴. However, in some lymphomas as an example, checkpoint blockade has been disappointing despite higher TMB. Hence additional factors may be present, such as low infiltration of immune cells or a tumor’s use of a checkpoint other than PD1 to evade the immune system. Therefore,, more comprehensive analysis of the immune microenvironment is needed to better match such hematologic malignancies with immunotherapy; such analyses may also help determine better therapies for the subgroup of solid cancers resistant to checkpoint blockade despite higher TMB and MSI-H.

PD-L1 amplification is another biomarker that can be associated with ICI response. In a study of 118,187 tumor samples from a deidentified database, the prevalence of PD-L1 amplification was 0.7%; the ORR for patients with solid tumors that harbored PD-L1 amplification and were treated with ICIs was 66.7% (6 of 9 patients with solid cancers).⁹² Response rates of 65% to 87% have been reported in patients with refractory classic Hodgkin lymphoma treated with ICIs and virtually all classic Hodgkin disease harbors PD-L1 amplification.^87–89 Taken together, the data suggest that biomarkers that predict responsiveness to immunotherapy can be observed in both hematologic and solid malignancies.

CAR T-Cells:

CAR T-cell therapy leverages the immune system’s T-cells to recognize and eradicate cancer cells. T-cells are isolated from patients and modified to target tumor-associated antigens. CAR-T therapy has achieved FDA approval for treating several blood cancers such as B-cell acute lymphoblastic leukemia, large B-cell lymphoma, and multiple myeloma by targeting CD-19 and B-cell maturation antigens.

CAR-T technology has faced greater challenges in solid cancers, including lack of reliable tumor-associated antigens, hypoxic cores, and immunosuppressive tumor environments. However, CAR T-cells are now being developed for solid cancers, and some targets appear in both liquid and solid malignancies. For instance, CD70 is a target for CAR T-cells in clinical trials. CD70 belongs to the tumor necrosis factor superfamily of molecules. It is aberrantly expressed in hematologic malignancies including non-Hodgkin lymphoma and AML, as well as in a number of solid tumors, including renal cell cancer, glioblastomas, osteosarcomas, thymic carcinoma, nasopharyngeal carcinoma, ovarian, lung, colon and pancreatic cancer and melanoma, with the highest expression in renal cell carcinomas (79.5%).⁹⁵ CAR T-cell clinical trials are now crossing the divide and targeting both hematologic and solid cancers, albeit in separate trials (NCT04438083; NCT05947487; NCT04502446).¹¹⁵

CHALLENGES FOR TISSUE-AGNOSTIC TRIALS/APPROVALS AND CROSSING THE SOLID/BLOOD DIVIDE

There are still important challenges for tissue-agnostic clinical trials and approvals. For example, molecular alterations that are therapeutically druggable in theory may be ineffective targets in practice if they are sub-clonal (present in only a fraction of the malignant cells) in a particular tumor or tumor type, or if a tumor harbors an alteration that acts as a stronger oncogenic driver (perhaps in a different pathway) than the putative target. Ideally, hematopoietic neoplasms should be included in basket trials only if the targeted alteration is an established driver of the disease, based on, for instance, pre-clinical evidence. There are also additional challenges that pertain specifically to bridging solid and hematologic cancers. For instance, there may be different tumor microenvironments in solid and hematologic cancers, and many hematologic malignancies already have therapies that are effective at inducing long-term remissions. Comprehensive molecular testing is also done more frequently/routinely in solid tumor patients. Finally, many patients with leukemias have low blood counts that would not meet the eligibility criteria for typical solid tumor trials; these exclusion criteria would need to be modified to accommodate specific hematologic malignancies.

CONCLUSIONS: BRIDGING THE GREAT DIVIDE

Tissue-agnostic FDA approvals have been established based on high response rates, often of remarkable durability. In practice, they provide drug access to patients with multiple types of rare and ultra-rare cancers bearing the appropriate biomarker.^116,117 Biomarker-based basket studies have been the cornerstone for tissue-agnostic approvals.¹¹⁸ The basic premise for these studies is that molecular alterations that function as drivers can be untethered from histology or organ of origin and that diverse cancers bearing these abnormalities can be vulnerable to matched targeted therapy. However, historically, these basket studies have had a clear divide between solid and blood cancers. Hence, most tissue-agnostic therapeutics focus on solid cancers. Even so, there is abundant evidence that both solid and hematologic cancers can carry the same driver molecular alterations, as well as identical actionable surface antigens, and that both groups can be responsive to matched targeted gene- and immune-based therapeutics. For instance, BRAF^V600E mutations, ALK, FGFR1 and NTRK fusions, IDH1/2 mutations, and PD-L1 amplification, are found in both solid and hematologic cancers and are sensitive to cognate inhibitors.^{23,85,92,99,100,119–122} Furthermore, specific antigens such as CD70, targetable by CAR T-cells, are also expressed on leukemias, lymphomas and diverse solid cancers. Therefore, the next generation of genomically driven basket studies should span both hematologic malignancies and solid tumors.