Considering molecular alterations as pan-cancer tissue-agnostic targets

Abstract

Owing to high response rates, the Food and Drug Administration has approved both gene- and immune-targeted drugs for tumor-agnostic, genomic biomarker-based indications, for lethal solid and blood cancers. We posit that current data support tissue-agnostic activity as a paradigm, rather than an exception to the rule.

It has been over 20 years since the BCR–ABL-targeted therapy imatinib changed chronic myelogenous leukemia from a fatal disease to one with a near-normal life expectancy¹. This and similar observations as well as sophisticated genome-analyzing techniques such as next-generation sequencing² have powered a revolution in oncology. Next-generation sequencing – the ‘molecular microscope’ – has identified numerous highly targetable cancer-driving genes including those that are exceptionally rare, occurring in less than 1% of all cancers^3,4. The genomic discoveries have led to the regulatory approvals of several tissue-agnostic gene- and immune-targeted drugs by the US Food and Drug Administration (FDA), mostly in solid tumors^5,6 (Table 1). Furthermore, there is a rich pipeline of compounds in development for newer tissue-agnostic molecular targets.

Table 1|.

Current tissue-agnostic genomic biomarker-based FDA approvals

Date of FDA approval	Drug(s)	Mechanism	Biomarker	Objective response rate	FDA indication	Hematological malignancies included in the FDA approval	Pediatrics included in the FDA approval (age range)	EMA tissue-agnostic approval	FDA approvals
23 May 2017	Pembrolizumab	Monoclonal antibody directed against PD-1	dMMR or MSI-H	~40% (median PFS not reached at median follow up of ~6 months)	Adult and pediatric patients with unresectable or metastatic, MSI-H or dMMR solid tumors that have progressed after previous treatment and who have no satisfactory alternative treatment options, or with MSI-H or dMMR colorectal cancer that has progressed after treatment with fluoropyrimidine, oxaliplatin and irinotecan	No	Yes (no limits)	No	https://go.nature.com/3MqmBUq
26 November 2018	Larotrectinib	Small molecule NTRK inhibitor	NTRK fusion	~75% (at 1 year, 71% of the responses ongoing and 55% of the patients remained progression-free)	Adult and pediatric patients with solid tumors that have an NTRK fusion without a known acquired resistance mutation (either metastatic or where surgical resection is likely to result in severe morbidity, and who have no satisfactory treatments or whose cancer has progressed after treatment)	No	Yes (no limits)	Yes	https://go.nature.com/3MsXikK
15 August 2019	Entrectinib	Small molecule NTRK inhibitor	NTRK fusion	~57% (median DOR of 10 months)	Adults and pediatric patients ≥12 years old with solid tumors that have an NTRK fusion without a known acquired resistance mutation, are metastatic or where surgical resection is likely to result in severe morbidity, and have progressed after treatment or have no satisfactory standard therapy	No	Yes (>12 years old)	Yes	https://go.nature.com/3Mt3tFj
16 June 2020	Pembrolizumab	Monoclonal antibody against PD-1	TMB ≥10 mutations per Mb	~29% (50% of patients had response durations ≥24 months)	Adult and pediatric patients with unresectable or metastatic high TMB (≥10 mutations per Mb) solid tumors, as determined by an FDA-approved test, that have progressed after previous treatment and who have no satisfactory alternative treatment options	No	Yes	No	https://go.nature.com/3FFBw9o
13 August 2021	Belzutifan	HIF-2a inhibitor	VHL germline alteration	~49% (depending on tumor type; 50% of patients had DOR >12 months)	Adult patients with von Hippel-Lindau syndrome who require therapy for associated renal cell carcinoma, central nervous system hemangioblastomas, or pancreatic neuroendocrine tumors, not requiring immediate surgery	No; approval is for limited number of indications	No	No	https://go.nature.com/46V7jzo
17 August 2021	Dostarlimab	Monoclonal antibody against PD-1	dMMR	~42% (median DOR was 34.7 months)	Adult patients with dMMR recurrent or advanced solid tumors, as determined by an FDA-approved test, that have progressed on or after previous treatment and who have no satisfactory alternative treatment options	No	No	No	https://go.nature.com/3QkNdaD
22 June 2022	Dabrafenib and trametinib	Small molecule inhibitors of BRAF and MEK, respectively	BRAF-V600E	~41% (median DOR > 10 months)	• Adult and pediatric patients ≥6 years of age with unresectable or metastatic solid tumors with BRAF-V600E mutation who have progressed after previous treatment and have no satisfactory alternative treatment options • Dabrafenib in combination with trametinib is not indicated for patients with colorectal cancer	No; this approval excludes colorectal cancer	Yes (>6 years old)	No	https://go.nature.com/46WkBvn
26 August 2022	Pemigatinib	Small molecule inhibitor thattargets FGFR1, FGFR2 and FGFR3	FGFR1 rearrangement	>70% complete cytogenetic response rate (median DOR not reached)	• Adults with relapsed or refractory myeloid/lymphoid neoplasms with FGFR1 rearrangement • Pemigatinib is also approved for cholangiocarcinoma with an FGFR2 fusion or rearrangement	Yes; hematological tumor-agnostic approval	No	No	https://go.nature.com/49kobkz; https://go.nature.com/40oHqWk
21 September 2022	Selpercatinib	Small molecule RET inhibitor	RET fusion	~44% (median DOR >20 months)	Adults with locally advanced or metastatic solid tumors with a RET gene fusion that have progressed on or after previous systemic treatment or who have no satisfactory alternative treatment options	No	No	No	https://go.nature.com/3sg9SNm

DOR, duration of response.

Tumor-agnostic approvals

Tumor-agnostic approvals started as a marriage between genomics and immunotherapy for solid malignancies. The first tumor-agnostic FDA approval was for pembrolizumab (a PD-1 inhibitor) in adult and pediatric mismatch repair deficient (dMMR) or microsatellite instability-high (MSI H) solid cancers; the approval was based on 149 patients from 5 clinical trials that included retrospectively analyzed data in which a 40% response rate was observed⁷. Pembrolizumab was also later approved for adult and pediatric patients with solid tumors bearing a tumor mutational burden (TMB) of at least 10 mutations per megabase, based on retrospective analysis of the KEYNOTE-158 trial, with a response rate of around 29%. In addition, in adults with dMMR solid tumors, the PD-1 inhibitor dostarlimab was approved (with a response rate of 42%)^5,6,8,9.

Among the tumor-agnostic therapeutic targets identified by next-generation sequencing are defects in mismatch repair genes, NTRK and RET alterations, BRAF mutations, FGFR1 rearrangement, and high TMB^8,9 (Table 1). Moreover, many of the responses are durable at 1,2 or more years of follow-up, especially in immunotherapy-treated patients (Table 1). These impressive outcomes highlight the use of applying targeted drugs to malignancies based on their genomic underpinnings and not simply their histological classification. There are also several emerging pharmacologically tractable pan-cancer tissue-agnostic targets, including but not limited to ERBB2, NRG1, ALK, RAS and KIT^8,9 (Table 2).

Table 2|.

Examples of potential tumor-agnostic biomarkers

Altered kinases (via mutations, fusions or amplifications)	DNA damage repair	Survival, growth, proliferation Ligands pathways	Other
ALK	BRCA	Homologous repair deficiency
EGFR	PI3K/AKT/mTOR pathway	MAPK pathway (such as KRAS-G12C)
ERBB2/HER2	NRG1 (ERBB3/4 ligand)	HGF (MET ligand)
KIT	Cell cycle signals	IDH1/2
MET

The listed pathways and genomic targets all have FDA-approved drugs and/or drugs in clinical trials and could be used to fashion a basket/platform trial and/or an N-of-1 combination trial to prove pan-cancer efficacy.

The genomically driven, tissue-agnostic approvals include agents that target both immune system abnormalities and aberrant genes. The median cross-tumor response rate for genomically driven tumor-agnostic approvals was 44% (Table 1). Most approvals are for adults, but some include children. Some approvals do not encompass all possible tissue types, whereas others do. Although many genomic alterations have similar therapeutic effects in solid and hematological malignancies, there is no tissue-agnostic approval that spans both.

Many of the tissue-agnostic approvals of gene-targeted therapies are in solid cancers. The NTRK inhibitors larotrectinib and entrectinib were both authorized for adult and pediatric subsets of patients with solid tumors containing an NTRK fusion, based on clinical trials showing response rates of between 57% and 75%^3,5,6,8,9. Next, in adults and children with BRAF-V600E mutations, except for colorectal cancer, the BRAF inhibitor dabrafenib and the MEK inhibitor trametinib were approved in combination, based on several cohort trials with a response rate of 41%. More recently, the RET inhibitor selpercatinib was approved in adults with RET-fusion cancers based on a multi-cohort trial with a response rate of 44% in various tumors other than non-small-cell lung cancer (NSCLC) and thyroid cancer, as well as data in RET-fusion-positive NSCLC (response rate of 64% (previously treated) to 85% (previously untreated)) and thyroid cancer (response rate of 70%)^5,8,9.

The European Medicine Agency (EMA) has also authorized a few of these agents – for example, the NTRK inhibitors larotrectinib and entrectinib – for NTRK fusion-bearing cancers as tissue-agnostic indications (Table 1).

In contrast to the situation for solid cancers, there is currently only one tissue-agnostic FDA approval for hematological malignancies, which is for pemigatinib (an FGFR1, FGFR2 and FGFR3 inhibitor) in adults with FGFR1-rearranged relapsed myeloid/Iymphoid neoplasms, based on a complete cytogenetic response rate of >70%¹⁰. Separately, pemigatinib is also approved by the FDA for FGFR2 fusion/rearrangement-bearing cholangiocarcinoma.

BRAF somatic mutation patterns in pan-cancer

Some mutations have different therapeutic implications in certain tissues. An apt example is BRAF-V600E mutations, which predict response to BRAF and MEK inhibitors in cancers as diverse as hairy cell leukemia (96% response rate to the BRAF inhibitor vemurafenib) to melanoma (50% response rate with vemurafenib)¹¹. By contrast, colorectal cancer is less responsive¹². In fact, for approval of the tissue-agnostic combined BRAF and MEF inhibitors dabrafenib–trametinib in solid tumors, the FDA excluded colorectal cancer because of the lower response rates.

A question that arises is whether resistance to BRAF and MEK inhibitors in colorectal cancer is because the colonic tissue itself bestows a resistance factor, or because secondary genomic pathways, such as EGFR signals, are activated in colorectal cancer. The fact that the BRAF inhibitor encorafenib together with the EGFR antibody etuximab have been approved by the FDA for colorectal cancer with BRAF-V600E mutations¹³ suggests the latter – that is, in colorectal cancer, co-activation of the EGFR pathway must be mitigated when addressing BRAF alterations.

Tumor-agnostic approvals for both solid and hematological malignancies

There has yet to be a tumor-agnostic approval for solid tumors that includes a hematological malignancy even though activity is seen in hematological malignancies. For example, BRAF-V600E alterations have been documented in almost all patients with hairy cell leukemia, as well as in some patients with non-Langerhans histiocytosis Erdheim–Chester disease and other hematological malignancies. Although not approved by the FDA, vemurafenib achieves response rates of almost 100% in patients with relapsed hairy cell leukemia¹⁴.

This question could also be asked in reverse. For example, the BCR–ABL fusion occurs almost exclusively in leukemias. However, a BCR–ABL fusion-bearing glioblastoma was recently reported and shown to be responsive to the BCR–ABL kinase inhibitor imatinib¹⁵. Similarly, IDH1 mutations occur in solid tumors and in hematological malignancies, and the FDA has approved the IDH1 inhibitor ivosidenib in IDH1-mutated acute myelogenous leukemia and cholangiocarcinoma (https://go.nature.com/3QI4DQ0).

Together, the data suggest that genomically driven basket studies in the pan-cancer tissue-agnostic setting spanning both hematological malignancies and solid tumors merit exploration.

Conclusions

There are a variety of established and emerging tissue-agnostic targets (Table 1) in development with robust evidence of activity for cognate targeted drugs against several tumor types. The benefit of tissue-agnostic approvals is that they provide the opportunity for drug access across many tissue types, including for patients with rare and ultra-rare cancers that are unlikely to have their own trials, especially in the setting of an infrequent or ultra-rare molecular alteration. Furthermore, there is increasing evidence that the molecular alterations in a tumor are the drivers of growth, rather than the organ site of origin, and this explains the high response rates seen with current tissue-agnostic approvals. The drawback to tissue-agnostic approvals is that there may be settings in which the approved drugs do not work well. However, this is true even for organ-of-origin-based approvals, in which it has long been accepted that there may be large subsets of patients who do not respond. Moreover, it is increasingly apparent that when there is resistance to a targeted drug despite the presence of the cognate aberrant target, the resistance is due to secondary molecular co-driver alterations rather than the tissue of origin itself. Furthermore, primary and secondary resistance in ‘responsive’ histologies may also be conferred by co-driver molecular alterations. Potential ways forward to further define responsive tumors should include multi-omic assessment of co-drivers and efforts to co-target those co-drivers to enhance responsiveness. In conclusion, we postulate that molecular alterations as pan-cancer targets are a paradigm rather than an exception–tissue of origin is not the issue.