Chapter 49: Targeted Therapy in Cancer

The Human Genome Project has enabled sequencing of human DNA and led to advancements in technologies that detect genomic, transcriptional, proteomic, and epigenetic changes. These technologies, combined with novel drug development, have accelerated the implementation of personalized medicine. Personalized medicine uses concepts of the genetic and environmental bases of disease to individualize prevention, diagnosis, and treatment (^1,2). Optimization of treatment using targeted therapy—molecules targeting specific enzymes, growth factor receptors, and signal transducers, thereby interfering with a variety of oncogenic cellular processes—and other strategies made possible by advances in translational medicine holds the promise of improving patient care (³).

This chapter focuses on targeted therapy in cancer therapeutics. The material is organized according to the key drivers of carcinogenesis in humans and summarizes the current state-of-the-art applications of personalized medicine.

Upregulation of the mitogen-activated protein kinase (MAPK) cascades RAF (rapidly accelerated fibrosarcoma) and MEK (MAPK/extracellular signal-regulated kinase [ERK]) contributes to carcinogenesis. Several cell surface molecules activate RAS (KRAS, NRAS, and HRAS), a family of guanine triphosphatases (GTPases) that activate downstream RAF protein kinases (BRAF, CRAF, and ARAF). The most important substrates of RAF kinases are MEK1 and MEK2 (MAPK/ERK kinases). The MEK kinases have one main substrate, ERK (⁴). Activation of ERK leads to modifications in gene expression mediated by transcription factors that control cell cycle progression, differentiation, metabolism, survival, migration, and invasion. This pathway regulates apoptosis by the posttranslational phosphorylation of apoptotic regulatory molecules (Bad, Bim, Mcl-1, and caspase 9). RAS is a downstream effector of the epidermal growth factor receptor (EGFR). Activation of ERK promotes upregulated expression of EGFR ligands and an autocrine loop critical for tumor growth (⁵). The frequency of molecular alterations in major pathway components is shown in the COSMIC (Catalogue of Somatic Mutations in Cancer) database (http://www.sanger.ac.uk/genetics/CGP/cosmic/).

Melanoma

Mutations in BRAF are found in 62% to 72% of patients with metastatic melanoma (⁶) and are less frequent in the radial growth phase (10%) and in situ (5.6%) melanomas (⁷). Mutations of NRAS occur in 5.2% of melanomas (⁷). In conjunctival melanoma, BRAF and NRAS mutations were identified in 29% and 18% of patients, respectively (⁸). Alterations of KIT were found in 36% and 39% of patients with acral and mucosal melanoma, respectively (⁹). Alterations of GNAQ and GNA11 were found in 45% and 32% of patients with uveal melanoma, respectively (¹⁰).

Inhibitors of BRAF and MEK have been approved by the US Food and Drug Administration (FDA) based on their significant antitumor activity and tolerability in patients with melanoma. The FDA-approved drugs and selected investigational agents by molecular target/pathway are listed in Table 49-1.

Vemurafenib and Dabrafenib

Vemurafenib was the first BRAF FDA-approved inhibitor for metastatic melanoma with a BRAF V600E mutation. A phase III trial demonstrated a 3.7-month improvement in progression-free survival (PFS) in the vemurafenib arm compared to the dacarbazine arm (median PFS 5.3 months and 1.6 months, respectively). The median overall survival (OS) was not reached in the vemurafenib arm and was 7.9 months in the control arm (¹¹). Dabrafenib is also FDA approved for patients with unresectable or metastatic melanoma with a BRAF V600E mutation, based on the results of a phase III study that compared dabrafenib with dacarbazine. The median PFS was 5.1 months and 2.7 months in the dabrafenib and the dacarbazine arms, respectively (¹²). Vemurafenib (¹³) and dabrafenib (¹⁴) have antitumor activity in patients with melanoma and brain metastases.

Trametinib

Trametinib is a MEK1/MEK2 kinase inhibitor that was approved by the FDA as a single agent or combined with dabrafenib for unresectable or metastatic melanoma with a BRAF V600E or V600K mutation. Approval was based on the results of a randomized trial, which demonstrated longer PFS with trametinib than with chemotherapy consisting of either dacarbazine or paclitaxel in patients with stage IIIc or IV melanoma and a BRAF V600E or V600K mutation (¹⁵). The median PFS durations were 4.8 and 1.5 months in the trametinib and chemotherapy arms, respectively (hazard ratio [HR], 0.47; P < .0001). The 6-month OS rates were 81% and 67%, respectively (¹⁵).

In a phase I and II study of dabrafenib plus trametinib or dabrafenib monotherapy in patients with melanoma and a BRAF V600E or V600K mutation, the objective response (complete response [CR] and partial response [PR]) rates were 76% and 54%, respectively (P = .03) (¹⁶). Cutaneous squamous cell carcinoma (SCC), an adverse event associated with BRAF inhibitors, was less common in the dabrafenib plus trametinib group than in the dabrafenib group (7% vs 19%, respectively) (¹⁶).

Other MEK inhibitors are in clinical trials. In a randomized phase II study in patients with BRAF-mutated advanced melanoma, selumetinib (MAP2K1/MAP2K2 inhibitor) plus dacarbazine was associated with longer PFS compared to dacarbazine (5.6 vs 3 months), but no improvement in OS was noted (¹⁷).

Lung Cancer

Mutations in BRAF occur in 1% to 4% of patients with non–small cell lung cancer (NSCLC). Molecular alterations in EGFR, ALK, ROS1, NRAS, and KRAS are also involved in the pathogenesis of lung cancer. We have noted responses in patients with NSCLC and BRAF V600E mutation treated with vemurafenib. A study of dabrafenib with or without trametinib in BRAF V600E–mutant NSCLC is ongoing.

Mutations of KRAS are more common in smokers. In metastatic NSCLC, mutated KRAS is associated with a worse prognosis than mutated EGFR. Mutation of KRAS was associated with shorter PFS in patients receiving maintenance erlotinib. No difference was noted in OS between mutated and wild-type KRAS (¹⁸). In colorectal cancer (CRC), KRAS mutations are associated with resistance to cetuximab. In a phase II study, selumetinib combined with docetaxel was associated with a higher response rate (37%, all PRs) in patients with KRAS-mutant NSCLC compared to docetaxel plus placebo (0%) (¹⁹). Other clinical trials evaluating MEK inhibitors combined with chemotherapy in KRAS-mutant NSCLC have been completed (NCT01192165, NCT01362296). In a phase II randomized study of trametinib compared to docetaxel in patients with KRAS-mutant NSCLC (²⁰), the rates of response and PFS were similar in the two arms (response, 12% [all PR]; median PFS, trametinib: 12 weeks; docetaxel: 11 weeks). Other studies of RAS-RAF-MEK inhibitors are ongoing.

The phosphatidylinositol 3-kinase (PI3K) signaling pathway is involved in the survival, growth, metabolism, motility, and progression of cancer and is a critical pathway in cancer (²¹). The PI3K family of proteins catalyzes the phosphorylation of phosphatidylinositols (PtdIns) at their 3’ position and consists of classes I, II, and III. Only class IA signaling aberrations are involved in human cancers (²²). The class IA PI3Ks are composed of heterodimers of regulatory subunits (p85α, p85β, p50α, p55α, and p55γ) and catalytic subunits (p110α, p110β, p110δ). Three genes encode the regulatory subunits: PIK3R1 encodes p85α (²²), and PIK3R2 and PIK3R3 encode the p85β and p55γ isoforms of the p85 regulatory subunit, respectively. Three genes, PIK3CA, PIK3CB, and PIK3CD, encode the highly homologous p110 catalytic subunit isoforms p110α, p110β, and p110δ and share a similar five-domain structure. At the amino terminus, there is an adapter-binding domain that interacts with the p85 regulatory subunit, followed by a RAS-binding domain that mediates interaction with RAS.

The class I PI3Ks can phosphorylate the 3’ position of PtdIns, PI-4-P, and PI-4,5-P₂ though PI-4,5-P₂. This phosphorylation generates the second-messenger phosphatidylinositol (3,4,5) triphosphate (PIP₃). Cytosolic proteins, such as the AKT family of protein-serine/threonine kinases, bind to PIP₃ and localize to the plasma membrane in response to PI3K activation (²³). In the absence of stimulated growth conditions, baseline levels of PIP₃ are undetectable in mammals. The PIP₃ levels at the plasma membrane are regulated by the tumor suppressor phosphatase and tensin (PTEN) homolog, whose lipid phosphatase activity converts PIP₃ to PI-4,5-P₂. Loss of PTEN function through inactivating mutations, deletion, chromosomal translocation, or epigenetic silencing is the second most common initiating event in cancer after p53 mutations.

Mutations or amplifications of the PI3K catalytic subunits p110α (PIK3CA) and p110β (PIK3CB), the PI3K regulatory subunits p85α (PIK3R1) and p85β (PIK3R2), and AKT (AKT1) can activate the PI3K pathway. Mutations, deletions, or epigenetic changes in negative regulators of the PI3K axis (PTEN and inositol polyphosphate-4-phosphatase, type II) may modify tumor cell sensitivity to chemotherapy or targeted therapies (²⁴). AKT is the main effector of PI3K activation and has three isoforms: AKT1, AKT2, and AKT3. AKT signaling plays a significant role in cell hypertrophy, survival, hyperplasia, and metabolism. Mammalian target of rapamycin (mTOR) is the catalytic subunit of mTOR complex 1 (mTORC1) and mTOR complex 2 (mTORC2), which are distinguished by their accessory proteins, regulatory-associated protein of mTOR (RAPTOR) and rapamycin-insensitive companion of mTOR (RICTOR) (²⁵).

Several studies focused on targeting the PI3K-AKT-mTOR pathway. Rapamycin analogues (rapalogs) have antitumor activity in various tumors and are frequently combined with other anticancer agents (²⁶). Everolimus is approved for the treatment of subependymal giant cell astrocytoma; hormone receptor (HR)-positive, HER2 (human epidermal growth factor receptor 2)-negative breast cancer (in combination with exemestane); neuroendocrine pancreatic tumors; tuberous sclerosis–associated subependymal giant cell astrocytoma; and renal cell carcinoma (sunitinib or sorafenib refractory). Temsirolimus is approved for renal cell carcinoma. The efficacy of rapalogs combined with endocrine therapy for advanced breast cancer was evident in the BOLERO-2 trial, which showed a median PFS of 6.9 months for everolimus and exemestane versus 2.8 months for exemestane alone (²⁷).

PI3K Inhibitors

The first generation of class I pan-PI3K inhibitors targeted PI3Kα, PI3Kβ, PI3Kγ, and PI3Kδ. Wortmannin and LY294002 had limited activity. Ongoing studies are evaluating new pan-PI3K inhibitors with improved pharmacokinetic profiles and target specificity. Their antitumor activity is primarily cytostatic. Novel agents that inhibit both PI3K and mTOR may improve the antitumor activity of either agent.

GDC-0941

GDC-0941 is a selective oral class I PI3K inhibitor and at high concentrations also an mTOR inhibitor. It is being investigated in clinical trials in patients with metastatic breast cancer (NCT00960960, NCT01437566) and advanced NSCLC. Initial studies of GDC-0941 demonstrated PRs in patients with melanoma and ovarian, cervical, and estrogen receptor (ER)–positive/HER-negative breast cancer (^28,29,30). At the maximum tolerated dose (MTD), dose-limiting toxicities (DLTs) included grade 3 macular rash and asymptomatic T-wave inversion on electrocardiograms, grade 3 thrombocytopenia, and grade 4 hyperglycemia (^29,30). Ongoing studies evaluating GDC-0941 include a phase II study in patients with untreated advanced or recurrent NSCLC treated with carboplatin/paclitaxel or carboplatin/paclitaxel/bevacizumab with or without GDC-0941 (NCT01493843). In a phase I/II study, GDC-0941 and cisplatin are being studied in patients with androgen receptor (AR)–negative, triple-negative, metastatic breast cancer, and in a phase II study, patients with advanced/metastatic breast cancer resistant to aromatase inhibitor therapy are being treated with GDC-0941 or GDC-0980 with fulvestrant versus fulvestrant alone (NCT01437566). Clinical trials are also investigating combinations of PI3K inhibitors taselisib (GDC-0032) or pictilisib (GDC-0941) with other targeted agents (eg, palbociclib, a cyclin-dependent kinase 4 and 6 [CDK4/6] inhibitor) in advanced solid tumors or breast cancer (NCT02389842).

BKM120

BKM120 is an oral pyrimidine-derived pan-PI3K inhibitor with activity against all class I PI3K isoforms. A phase I study demonstrated that BKM120 was well tolerated, with a dose-dependent safety profile (³¹). Adverse events included hyperglycemia, rash, nausea, fatigue, and mood alterations. Hyperglycemia is a typical adverse event associated with the use of PI3K/AKT/mTOR pathway inhibitors. Another study demonstrated that in colorectal, breast, lung, and endometrial cancers treated with BKM120, two of 77 patients had a PR (triple-negative breast cancer with KRAS and p53 mutations, n = 1; and ER-positive/HER-negative metastatic breast cancer, n = 1; both had tumor PIK3CA mutations), and 58% of patients had stable disease (SD).

BAY 80-6946

BAY 80-6946 is a pan–class I PI3K inhibitor with activity against PI3Kα, PI3Kβ, PI3Kδ, and PI3Kγ. A phase I study demonstrated that the MTD of BAY 80-6946 was 0.8 mg/kg intravenously weekly (3 weeks on, 1 week off). Adverse events included hyperglycemia, fatigue, nausea, diarrhea, and mucositis. Clinical benefit was reported in patients with advanced breast, endometrial, and gastric cancers.

BEZ235

BEZ235 is an oral, reversible, and selective inhibitor of PI3K and TORC1/2. Preclinical data demonstrated antitumor activity in melanoma, breast, CRC and sarcoma. BEZ235 suppresses cell proliferation, induces G1 cell cycle arrest, and promotes autophagy by inhibiting the activity of AKT, S6K, S6, and 4EBP1 target proteins. BEZ235 has been investigated in phase I/II clinical trials in patients with advanced cancer alone (³²) or in combination with paclitaxel, trastuzumab, everolimus, or MEK162. In a phase IB study, BEZ235 combined with trastuzumab in 15 patients with HER2-positive metastatic breast cancer with altered PI3K/PTEN status was tolerable. Stable disease and PR were reported in four and one patients, respectively (³³). An improved formulation of BEZ235 was used as a monotherapy or combined with trastuzumab, and SD was noted in 40% of patients with advanced cancer. The most common adverse events were nausea, diarrhea, elevated transaminases, and headache. The DLTs were fatigue, asthenia, grade 3 thrombocytopenia, and grade 3 mucositis (³²). A clinical trial of BEZ235 and everolimus in advanced cancer is ongoing (NCT01628913).

In prostate cancer, PTEN loss may be associated with resistance to castration (^34,35). BEZ235 causes growth arrest in PTEN-negative prostate cancers, but inhibition of the PI3K pathway leads to activation of AR signaling (inhibition of AR appears to result in promotion of PI3K activity) (³⁵).

Other p110α Isoform-Specific Inhibitors

Other p110α isoform-specific inhibitors, such as BYL719, GDC-0032, and INK1117, are being investigated in various solid tumors. BYL719 was associated with less hyperglycemia than the pan-PI3K inhibitor BKM120 (³⁶). In a phase I study, BYL719 induced tumor reduction in 33% of patients with ER-positive, metastatic breast cancer and a PIK3CA mutation (³⁷). Multiple studies are investigating the role of BYL719 in solid tumors as a single agent or in combination with targeted agents and cytotoxics. Although preclinical data demonstrated that PIK3CA alterations are the best biomarkers for predicting sensitivity to BYL719, p110α inhibitors are not effective in PIK3CA-mutated cells that also have a PTEN deletion (³⁷).

AKT Inhibitors

The AKT inhibitors may induce the PI3K-stimulating receptor tyrosine kinase HER3 in breast cancer cell lines and may increase IGF-1R and the insulin receptor, thereby leading to the development of escape pathways and resistance mechanisms. Combination therapies that block the feedback response may overcome resistance to AKT inhibitors. Several studies have investigated or are investigating AKT inhibitors (such as MK2206, GSK2141795, and BAY1125976) as single agents or in combination with targeted therapies or chemotherapy in specific tumor types (examples are NCT01333475, NCT01902173, NCT01979523, and NCT01915576). Other drugs, such as MSC2363318A, a dual p70S6K/AKT inhibitor, are in clinical trials (NCT01971515).

Based on the increase in tumor inhibition with combined MEK/PI3K targeting and the tolerability of drugs targeting each pathway individually, early-phase trials combining GDC-0941 (PI3K inhibitor) with GDC-0973 (MEK inhibitor) and combining BKM120 (PI3K inhibitor) with GSK1120212 (MEK inhibitor) have been completed (^38,39). The latter study demonstrated promising antitumor activity in patients with KRAS-mutant ovarian cancer.

Conclusions

In summary, molecular alterations in the PI3K/AKT/mTOR pathway have been identified in multiple tumor types, emphasizing the critical role of this pathway in tumorigenesis and disease progression. The PI3K inhibitors, as single agents, have mostly cytostatic activity. Several escape mechanisms are involved in resistance to PI3K/AKT/mTOR inhibitors. Clinical trials are exploring the role of PI3K, AKT, or mTOR inhibitors in combination with other targeted or cytotoxic agents. In our experience, patients with molecular alterations in the PI3K/AKT/mTOR pathway treated with targeted therapies have shorter survival compared to patients with alterations in the RAS/RAF/MEK or EGFR/HER/other pathways treated with the matched targeted agents, perhaps due to less-effective therapies than those for other pathways or intrinsic resistance (unpublished data). Carefully designed clinical trials, patient selection, and the elucidation of mechanisms of response and resistance to PI3K/AKT/mTOR pathway inhibitors, including protein and phosphoprotein expression with signatures of sensitivity/resistance, may improve clinical outcomes.

Epidermal growth factor receptor (ErbB1, HER1) is a cell surface transmembrane receptor that belongs to the EGF family of extracellular protein ligands. It is a member of the ErbB receptor family, which consists of four receptor tyrosine kinases: EGFR, HER2 (ErbB2), HER3 (ErbB3), and HER4 (ErbB4). The EGF binds to EGFR, stimulating ligand-induced dimerization, receptor dimerization, and signaling through tyrosine kinase activity, leading to activation of multiple pathways involved in cell proliferation, survival, metastases, and neoangiogenesis (⁴⁰).

Epidermal Growth Factor Receptor–Targeted Therapies

Therapies that target EGFR include tyrosine kinase inhibitors (TKIs) and monoclonal antibodies. Gefitinib was the first selective EGFR inhibitor and was approved by the FDA in 2003 for NSCLC, but in 2005 the FDA withdrew this approval for use in new patients because gefitinib did not improve survival compared to placebo in previously treated patients (⁴¹).

Erlotinib

Erlotinib targets the EGFR tyrosine kinase and is approved by the FDA for first-line treatment of patients with metastatic NSCLC with tumor EGFR exon 19 deletions or exon 21 (L858R) substitution mutations; maintenance therapy of patients with NSCLC and no evidence of disease progression after four cycles of platinum-based first-line chemotherapy; and treatment of NSCLC after failure of one or more prior chemotherapy regimens. It is approved as first-line treatment, in combination with gemcitabine, of patients with locally advanced/metastatic pancreatic cancer. In NSCLC, erlotinib demonstrated a significant improvement in median PFS and OS compared to placebo (⁴²). Mutations in the EGFR kinase domain predicted response to EGFR TKIs (^43,44). In patients with EFGR alterations, the response rate to EGFR TKIs ranged from 48% to 90% (^45,46). Randomized trials demonstrated that the use of gefitinib or erlotinib is associated with longer PFS compared to platinum doublets in patients with lung adenocarcinoma and activating EGFR mutations. However, no OS benefit was noted (^{47,48,49,50,51}), perhaps partially due to crossover after disease progression.

In a randomized study of erlotinib versus placebo as maintenance therapy in patients with advanced NSCLC who had an objective response or SD after four cycles of a platinum-based doublet, erlotinib was associated with superior PFS in patients with adenocarcinoma or SCC. Survival benefit was noted only in patients with SCC (⁵²). The role of EGFR TKIs in patients with wild-type EGFR is unclear (⁵³).

The role of adjuvant erlotinib in patients with resected NSCLC and EGFR molecular alterations was investigated in a phase III trial (comparing placebo vs erlotinib) (NCT00373425) (⁵⁴). In 973 randomized patients, there was no difference in disease-free survival between the two arms (HR, 0.90; P = .32). In a subset analysis of 161 patients with deletion 19 or L858R EGFR mutations, the median disease-free survival was 46.4 months in the erlotinib arm versus 28.5 months in the placebo arm (HR, 0.61; P = .04) (^55,56). In a phase II study (RTOG 1306, NCT01822496), patients with stage III EGFR-mutant lung cancer or ALK-positive NSCLC were randomized to erlotinib followed by concurrent chemoradiation, crizotinib followed by concurrent chemoradiation, or chemoradiation alone.

Afatinib

Newer agents irreversibly inhibit EGFR and target additional EGFR members, such as HER2 and HER4. Afatinib is a selective, oral inhibitor of EGFR/ErbB1, HER2, and HER4. In a phase III study of afatinib or cisplatin plus pemetrexed in patients with metastatic lung adenocarcinoma with EGFR mutations, afatinib was associated with longer PFS compared to standard doublet chemotherapy (11.1 months vs 6.9 months, respectively; P = .001) (⁵⁷). In patients with exon 19 deletions or L858R mutations, the median PFS was 13.6 months. Development of resistance is attributed to additional EGFR mutations (T790M on exon 20, 50% of patients) (⁵⁸) or PIK3CA mutations; MET or HER2 amplification; epithelial-to-mesenchymal transformation; or transformation to small cell lung cancer. Novel third-generation EGFR TKIs that target T790M mutations such as CO-1686 (rociletinib) are being investigated.

Cetuximab

Cetuximab is a chimeric mouse-human immunoglobulin 1 monoclonal antibody against EGFR. It is indicated for the treatment of advanced SCC of the head and neck (combined with radiation therapy); recurrent or metastatic SCC of the head and neck (combined with platinum-based therapy with 5-fluorouracil); and recurrent or metastatic SCC of the head and neck that progresses after platinum-based therapy. Cetuximab with FOLFIRI (irinotecan, fluorouracil, and leucovorin) is indicated for the first-line treatment of K-RAS wild-type, EGFR-positive metastatic CRC and cetuximab with irinotecan for irinotecan-refractory patients; cetuximab monotherapy is indicated for patients with oxaliplatin- and irinotecan-refractory CRC or patients with irinotecan intolerance.

The role of cetuximab in lung cancer has not been determined. In a phase III study, the addition of cetuximab to taxane/carboplatin did not significantly improve PFS or OS (⁵⁹). In another phase III study in patients with advanced EGFR-expressing NSCLC, the addition of cetuximab to cisplatin and vinorelbine was associated with improved OS compared to the cisplatin and vinorelbine arm (median OS, 11.3 months vs 10.1 months, respectively; P = .04) (⁶⁰). The SWOG 0819 trial comparing carboplatin/paclitaxel with and without bevacizumab (in eligible patients) and/or cetuximab in stage IV or recurrent NSCLC has been suspended (NCT00946712).

Mutations and overexpression of EGFR are frequent in patients with CRC (⁶¹). In patients with irinotecan-refractory CRC, cetuximab demonstrated significant antitumor activity as monotherapy (overall response rate [ORR], 10.8%; median time to progression, 1.5 months; median OS, 6.9 months) or combined with irinotecan (22.9%; 4.1 months and 8.6 months, respectively) (⁶¹). In a phase III study, cetuximab was associated with improved survival compared to best supportive care in patients with refractory metastatic CRC (6.1 months vs 4.6 months) (⁶²). In another phase III study, first-line treatment with cetuximab plus FOLFIRI reduced the risk of progression compared to FOLFIRI alone in patients with metastatic CRC, but the benefit of cetuximab was limited to patients with KRAS wild-type tumors (⁶³).

In patients with previously untreated metastatic CRC, cetuximab plus FOLFOX-4 (oxaliplatin, leucovorin, and fluorouracil) was associated with a higher ORR (43% vs 36%) compared to FOLFOX-4 alone (⁶⁴). In patients with KRAS wild-type CRC, cetuximab plus FOLFOX-4 compared to FOLFOX-4 alone was associated with a higher ORR (61% vs 37%; P = .01) and a lower risk of disease progression (HR, 0.57; P = .02). This study demonstrated that KRAS mutational status is a highly predictive selection criterion for the addition of cetuximab to FOLFOX-4 in this setting (⁶⁴).

Panitumumab

Panitumumab is a fully humanized IgG2 anti-hEGFR antibody indicated for the treatment of patients with wild-type KRAS (exon 2 in codons 12 or 13) metastatic CRC as first-line therapy in combination with FOLFOX and monotherapy in fluoropyrimidine-, oxaliplatin-, and irinotecan-refractory patients. In a phase III study of panitumumab with FOLFOX4 versus FOLFOX4 alone as first-line therapy in patients with wild-type KRAS metastatic CRC, the addition of panitumumab to FOLFOX4 resulted in higher rates of overall response (57% vs 48%; P = .02), PFS (10 vs 8.6 months; P = .01), and OS (HR, 0.83; P = .03) (⁶⁵). In a randomized study of panitumumab versus cetuximab in patients with chemotherapy-refractory wild-type KRAS exon 2 metastatic CRC, panitumumab was not inferior to cetuximab, and the OS benefit was similar (median OS, 10.4 vs 10 months, respectively; HR, 0.97) (⁶⁶).

The tyrosine-protein kinase KIT (c-Kit or CD117) is a receptor tyrosine kinase protein that is encoded by the KIT gene. KIT regulates cell differentiation and proliferation, resists cell apoptosis, and plays an important role in tumorigenesis and migration by activating downstream signaling molecules following interaction with stem cell factor (SCF). Complete loss of KIT activity results in in utero or perinatal death; “loss of function” leads to failure of particular stem cell populations to migrate and survive. Activating mutations of KIT occur in almost all patients with systemic mastocytosis. In gastrointestinal stromal tumors (GISTs), activating mutations of KIT occur in more than 80% of patients; two-thirds of KIT mutations occur in exon 11, resulting in dysfunction of the intracellular autoinhibitory juxtamembrane domain. The majority of these mutations are indels. Deletion of exon 11 is associated with shorter PFS and OS in patients with GIST (⁶⁷). Approximately 10% to 15% of KIT mutations in GIST occur in the extracellular region encoded by exon 9 (primarily in intestinal GIST). Mutations in KIT occur in core binding factor leukemias (17% of acute myeloid leukemia [AML]), in up to 26% of patients with testicular seminomas, and in 30% of patients with unilateral ovarian dysgerminomas (⁶⁸). Activating KIT mutations and amplifications have been reported in 5% of patients with melanoma. In melanoma, KIT mutations occur as follows: exon 9 (5%), exon 11 (45%), exon 13 (25%), exon 17 (10%), and exon 18 (15%); more than 90% of these mutations are missense mutations. Therefore, KIT inhibition is an attractive therapeutic strategy in patients with tumor aberrations of the SCF/c-KIT signaling pathway.

Gastrointestinal Stromal Tumor

Imatinib binds directly to the ATP-binding site within KIT, competitively inhibiting ATP binding and stabilizing the kinase in the inactive conformation. In the preimatinib era, the median survival of patients with advanced GIST was less than 1.5 years. The use of imatinib in advanced GIST is associated with a median survival of approximately 5 years (⁶⁹). In the adjuvant setting, imatinib was associated with decreased risk of relapse after surgery with curative intent. The reintroduction of imatinib after disease relapse is associated with results inferior to those for continued imatinib therapy, suggesting that imatinib should not be interrupted (⁶⁹). Primary resistance to imatinib is noted in 10% of patients with GIST and is attributed to the type of KIT mutation (exon 9 mutations have a threefold higher risk of resistance than exon 11 mutations) or to suboptimal dose in patients with KIT exon 9 mutations (⁷⁰). Secondary resistance is attributed to acquired mutations in the ATP-binding site (exons 13/14) that interfere with imatinib binding or in the activation loop (exons 17/18) that stabilize the active conformation of KIT (⁷¹). In patients with imatinib-resistant GIST tumors, sunitinib (TKI/anti-VEGF agent) demonstrated significant improvement in the median time to progression compared to placebo, leading to FDA approval for this indication (⁷²). The efficacy of sunitinib in imatinib-secondary resistance is attributed to structural and enzymatic characteristics of sunitinib. Other tyrosine kinases targeted by sunitinib may play a role in its efficacy.

Regorafenib is FDA approved for TKI-resistant GIST. In a phase III study, regorafenib was associated with longer PFS than placebo (median PFS, 4.8 months vs 0.9 months, respectively; P < .0001) (⁷³). Although antitumor activity was noted with dasatinib and sorafenib, the use of nilotinib has not shown significant antitumor activity in the third-line setting (⁷⁴).

Inhibitors of KIT combined with other targeted agents may improve the outcomes of patients with KIT alterations and overcome resistance. A phase I study of dasatinib and ipilimumab for unresectable or metastatic GIST or other sarcomas is ongoing (NCT01643278).

Melanoma

In patients with melanoma and a KIT alteration, the use of imatinib was associated with an ORR of 23.3% (⁷⁵). Patients whose disease responds to imatinib typically have activating mutations in exons 11 (L576p; response, 64%) and 13 (K462E; response, 43%). Other mutations in exon 11 (eg, V559X, V560D) are also sensitive to imatinib. Wild-type KIT amplification is not sensitive to imatinib. In smaller studies, sunitinib or nilotinib also had antitumor activity (^76,77).

Mastocytosis

One of the indications of imatinib is the treatment of adults with systemic mastocytosis and a non-KIT D816V mutation (or unknown KIT mutation status). A KIT D816V mutation, which occurs in most patients with mastocytosis, is resistant to imatinib. In a phase II clinical trial, dasatinib was associated with symptomatic benefit in patients with mastocytosis harboring a KIT D816V mutation (⁷⁸). In this setting, responses have been reported with midostaurin (NCT00782067).

As previously discussed, the HER family includes four receptors: HER1, HER2, HER3, and HER4. HER2 is involved in the regulation of proliferation and survival of epithelial cells and is considered an orphan receptor because it has no known ligand. HER1, HER3, and HER4 receptors have ligands and form homodimers or heterodimers on ligand binding. HER2 can heterodimerize with any of the other receptors and is the preferred dimerization partner, leading to autophosphorylation of tyrosine residues within the cytoplasmic domain of the receptors and initiating signal transduction via the PI3K/AKT and RAS/MAPK pathways (⁷⁹). In breast cancer, amplification and overexpression of the HER2 oncogene is a poor prognostic factor (⁸⁰). Overexpression of HER2 occurs in approximately 15% to 20% of early-stage breast cancer and was historically associated with poor clinical outcomes.

Studies combining HER2-targeted therapies with other agents (PI3K, mTOR inhibitors) or immunotherapy and combinations of two or more HER2-targeted therapies (trastuzumab and TDM1; trastuzumab and pertuzumab followed by TDM1) have been completed or are ongoing (examples are NCT02073487, NCT02073916, NCT01835236, and NCT02252887).

Trastuzumab

Trastuzumab is a recombinant DNA-derived humanized monoclonal antibody that selectively binds with high affinity to the extracellular domain of HER2 protein. The antibody is an IgG1 kappa that contains human framework regions with the complementarity-determining regions of a murine antibody that binds to HER2. Trastuzumab is indicated as monotherapy for patients with metastatic breast cancer whose tumors overexpress HER2 protein and had one or more chemotherapy regimen(s) and combined with paclitaxel for patients with metastatic breast cancer whose tumors overexpress HER2 protein and who have not received chemotherapy for their metastatic disease. Approximately 15% of patients develop disease recurrence after trastuzumab therapy. Resistance to trastuzumab has been attributed to altered receptor-antibody interaction, activation of the downstream pathways by increased signaling from other members of the HER family or other receptors, or constitutive activation of downstream elements. Prospective studies are assessing the role of trastuzumab in patients whose breast cancers do not overexpress HER2.

Lapatinib

Lapatinib, a dual TKI of EGFR and HER2, has antitumor activity in trastuzumab-refractory breast cancer. Lapitinib is FDA approved in combination with capecitabine for patients with advanced or metastatic breast cancer whose tumors overexpress HER2 and who have received prior therapy, including anthracycline, taxane, and trastuzumab therapy, and letrozole for postmenopausal women with hormone receptor-positive metastatic breast cancer that overexpresses the HER2 receptor and for whom hormonal therapy is indicated.

Trastuzumab Emtansine

Trastuzumab emtansine (ado-trastuzumab emtansine, T-DM1) is an antibody-drug conjugate consisting of trastuzumab linked to the cytotoxic agent mertansine (DM1). Trastuzumab inhibits the growth of cancer cells by binding to HER2, and mertansine enters and destroys cells by binding to tubulin. T-DM1 is specifically toxic to tumor cells because it targets HER2, which is overexpressed in cancer cells. In patients with HER2-positive, trastuzumab-resistant breast cancer, T-DM1 improved survival by 5.8 months compared to lapatinib and capecitabine combination therapy (⁸¹). This study led to the FDA approval of T-DM1, which is indicated as monotherapy for patients with HER2-positive, metastatic breast cancer previously treated with trastuzumab with or without a taxane. Patients receiving T-DM1 should have received prior therapy for metastatic disease or developed disease recurrence during or within 6 months of completing adjuvant therapy.

Pertuzumab

Pertuzumab is a humanized monoclonal antibody that binds to the HER2 receptor and inhibits the interaction between HER2 and other HER family members (HER1, HER3, and HER4) on the surface of cancer cells. Pertuzumab is FDA approved in combination with trastuzumab and docetaxel as a neoadjuvant treatment for patients with HER2-positive breast cancer and as a treatment for patients with HER2-positive metastatic breast cancer. Clinical trials are evaluating the role of pertuzumab or T-DM1, combined with standard chemotherapy and trastuzumab, for early-stage breast cancer in the adjuvant setting.

In many cancer cells, CDK4 and CDK6 mediate cell cycle control. Randomized phase II trials in patients with ER-positive metastatic breast cancer demonstrated that CDK4/6 inhibition combined with first-line endocrine treatment can improve PFS. Clinical trials with CDK4/6 inhibitors as single agents or combined with other drugs are ongoing.

The AR is expressed in the vast majority of ER-positive breast cancers. Interestingly, a subset of ER-negative tumors also expresses AR. Ongoing clinical trials will define the role of androgen deprivation and AR blockade in ER-positive and ER-negative tumors.

Hereditary BRCA1 and BRCA2 mutations are associated with an increased risk of developing breast and ovarian cancer. BRCA1/2 mutations lead to impaired double-strand DNA repair. PARP1 plays an important role in repairing single-strand breaks, and PARP1 inhibitors result in the formation of multiple double-strand breaks. Thus, the use of PARP inhibitors in tumors with BRCA1, BRCA2, or PALB2 mutations renders the cells unable to efficiently repair DNA and makes them vulnerable to apoptosis. In phase II studies in BRCA-associated breast cancer, the use of PARP inhibitors was associated with clinical responses (⁸²). Ongoing studies are comparing PARP inhibitors with standard chemotherapy in women with BRCA1/2-mutated advanced breast cancer. Clinical trials of PARP inhibitors in patients with deleterious BRCA1/2 mutations have been completed or are ongoing in breast and ovarian cancer and in other tumor types (examples are NCT01989546, NCT02326844). The PARP inhibitors in clinical testing include olaparib (as a single agent or in combination with the PI3K inhibitors BKM120 or BYL713), veliparib, rucaparib, and talazoparib as single agents or in combination with cytotoxics.

c-MET (or MET) is a receptor tyrosine kinase with specificity for a single ligand, the hepatocyte growth factor (HGF). Binding of HGF to MET leads to receptor dimerization and autophosphorylation of MET on its intracellular kinase domain and subsequent phosphorylation of its C-terminal docking site and juxtamembrane domain. These phosphorylation events enable activation of multiple downstream effector proteins, such as the adaptor proteins Grb2 and Gab1, leading to activation of the PI3K, Ras/RAF/MEK/ERK, PLC-γ, STATs, and FAK signaling pathways. c-MET plays a role in promoting the proliferation, survival, motility, and invasion of normal and tumor cells. It is thought c-MET promotes metastasis through increased production of HGF by hepatocytes, leading to enhanced paracrine signaling and clonal selection of metastatic cells with high MET expression (^83,84). c-MET promotes angiogenesis, and it is involved in the development of resistance to cytotoxic chemotherapy and VEGF or EGF receptor inhibitors. c-MET and phospho-c-MET have been associated with poor clinical outcomes in patients with CRC and lung cancer and with disease progression in patients with breast cancer and melanoma. The intracellular pathways activated by c-MET interact with receptors such as EGFR, HER2, WNT, and the insulinlike growth factor receptor 1 (IGFR1). Activation of EGFR leads to c-MET phosphorylation and activation. Activation of the WNT-β-catenin pathway results in MET transcription (⁸⁵). c-MET activation can promote other growth receptor pathways, including the HER3-PI3K-AKT signaling pathway (⁸⁶). The c-MET mutations are usually a result of sporadic somatic alterations acquired during cancer development (^83,84,86,87). Germline MET mutations are found in papillary renal cell carcinoma, familial gastric cancer, and CRC. Development of resistance to sunitinib (⁸⁸), gefitinib (⁸⁶), and erlotinib (⁸⁷) has been in part attributed to c-MET activation. Paired analysis of tumors from patients with lung cancer whose disease stopped responding to gefitinib or erlotinib identified MET amplification as an acquired resistance mechanism in selected patients (⁸⁶).

Targeted therapy against the HGF/MET pathway includes small-molecule inhibitors and antibodies against either HGF or MET. Cabozantinib is a small-molecule inhibitor that inhibits c-Met and VEGFR2 and is FDA approved for medullary thyroid cancer. Monoclonal antibodies include rilotumumab and ficlatuzumab. Rilotumumab was in development, but all clinical trials in advanced gastric cancer (including two phase III studies) were closed after a randomized trial demonstrated an increased death rate in the rilotumumab and chemotherapy combination arm compared to the chemotherapy arm. A phase II randomized, double-blind study is comparing ficlatuzumab plus erlotinib with placebo plus erlotinib in patients with previously untreated, metastatic, EGFR-mutated NSCLC and “BDX004-positive label” (NCT02318368). Onartuzumab is a monovalent humanized monoclonal antibody produced in Escherichia coli that binds to the Sema domain on the extracellular part of MET to block HGF binding. A phase III study that compared onartuzumab plus erlotinib versus erlotinib plus placebo in patients with MET-positive advanced NSCLC was discontinued owing to the lack of clinically meaningful efficacy in an interim analysis.

Clinical trials have been completed or are in development for the monoclonal antibodies against c-MET, such as LY-2875358, h224G11A, and DN30. Synthetic small-molecule unselective (crizotinib, foretinib, cabozatinib, MGCD-265) and selective (tivantinib, JNJ-38877605, AMG337, AMG208, PF-04217903, EMD-1214063, LY-2801653, and INC-280) c-MET TKIs are being investigated. A phase III trial of tivantinib plus erlotinib for the treatment of patients with locally advanced or metastatic, nonsquamous NSCLC was discontinued because it did not meet its primary end point of prolonging OS.

Gene rearrangements of ALK occur in 2% to 7% of patients with NSCLC. In these patients, the use of the ALK inhibitor crizotinib as first-line therapy was associated with longer PFS and less toxicity compared to chemotherapy (^89,90). Crizotinib is a small-molecule inhibitor initially developed to target c-MET (and is highly specific in this targeting) that is FDA approved for the treatment of ALK-positive NSCLC. A phase II trial is evaluating the role of crizotinib in predefined tumor types in patients whose tumors are harboring specific alterations in ALK or MET (NCT01524926). The next-generation ALK inhibitors were developed to overcome resistance to crizotinib, which is attributed to secondary mutations within the ALK-TK domain; EML4-ALK amplification; bypass activation of alternative signaling pathways (EGFR, c-KIT, KRAS, IGF1 receptor); and progression/occurrence of central nervous system (CNS) metastases.

Ceritinib

Ceritinib is an ATP-competitive ALK inhibitor that showed greater antitumor potency than crizotinib in preclinical studies. Ceritinib is FDA approved (accelerated process) for patients with ALK-positive NSCLC previously treated with crizotinib. In a phase I study in patients with ALK-rearranged NSCLC (n = 114) who had received 400 mg or more of ceritinib, the response rate was 58%, and the median PFS was 7 months. The response rate for crizotinib-pretreated patients was 56%. The median PFS was longer in crizotinib-naïve (10.4 months) than in crizotinib-pretreated (6.9 months) patients (⁹¹). In an expansion cohort of 246 patients with ALK-rearranged metastatic NSCLC treated with the MTD of 750 mg orally daily, the response rate was 58.5% (ALK inhibitor pretreated, 54.6%; ALK inhibitor naïve, 66.3%), and the median PFS was 8.2 months (6.9 months and not reached, respectively). Central nervous system responses were noted in some patients with untreated lesions.

A randomized trial for patients with ALK-rearranged metastatic NSCLC previously untreated or treated with chemotherapy and crizotinib is comparing ceritinib with pemetrexed/cisplatin or pemetrexed/carboplatin (NCT01828099). A phase III trial is comparing ceritinib with standard chemotherapy in patients with ALK-rearranged NSCLC. Other studies are investigating ceritinib as monotherapy in cholangiocarcinoma with ROS1 or ALK overexpression or in anaplastic/undifferentiated thyroid cancer with ALK alterations or in combination with targeted therapies or chemotherapy, such as with everolimus, in solid tumors and ALK-rearranged NSCLC, with chemotherapy in pancreatic cancer, with LEE011 in ALK-positive NSCLC, and with nivolumab in NSCLC.

Alectinib

Alectinib is another second-generation ALK inhibitor. In a phase I/II study of alectinib in ALK inhibitor-naïve patients with metastatic ALK-rearranged NSCLC, no DLTs were noted up to doses of 300 mg twice daily, and the response rate was 93.5%. In another phase I/II study in patients with ALK-rearranged tumors who had progressed on crizotinib treatment, the MTD of alectinib was 600 mg twice daily. The response rate during the dose escalation part of the study was 55%. In patients with known CNS metastases treated with alectinib, the response rate was 52% within the CNS. No patient initially free of CNS metastases developed new CNS lesions during treatment. A phase III trial is comparing alectinib to crizotinib in treatment-naïve patients with advanced ALK-rearranged NSCLC.

Other ALK Inhibitors

Another ALK inhibitor with antitumor activity in NSCLC is AP26113, which has activity against ALK and mutant isoforms of EGFR without activity against wild-type EGFR. Preliminary data from a phase I/II trial in crizotinib-pretreated patients demonstrated a response rate of 63%. ASP3026 is a selective, ATP-competitive ALK inhibitor with activity against wild-type ALK and against the most frequent crizotinib resistance-mediating gatekeeper mutation, L1196 M. In a phase I trial, the MTD was 125 mg, and 44% of patients with crizotinib-resistant ALK-rearranged NSCLC had a PR. TSR-011 is an inhibitor with antitumor activity against ALK and tropomyosin-related kinase. PD-06463922 is a TKI with activity against ALK and ROS1, which appears to retain its signal-blocking effects even when several ALK mutations, leading to crizotinib resistance, are present. X-396 is a small-molecule TKI blocking ALK signaling in the wild-type conformation and with considerable activity against at least two of the crizotinib resistance-mediating ALK point mutations (L1196 M and C1156Y). An ongoing phase I study demonstrated promising activity (NCT01625234). The heat shock protein 90 (HSP90) inhibitors have also demonstrated activity against ALK-rearranged NSCLC in early clinical trials. In a phase II trial of the HSP90 inhibitor ganetespib, PRs were noted in 4 of 98 patients (⁹²). The HSP90-inhibiting compound AUY922 has shown promising activity.

Notch was identified as an oncogene in T-cell acute lymphoblastic leukemia (T-ALL) in which the (7;9) chromosomal translocation fuses the N-terminal region of the T-cell receptor beta (TCRβ) to the C-terminus of Notch1. This leads to expression of a truncated Notch1 protein that lacks the extracellular subunit and is thus constitutively active (⁹³). The intracellular forms of all four Notch proteins are potentially oncogenic and capable of transforming normal cells. Deregulated expression of Notch proteins, ligands, and targets has been described in various solid tumors, including cervical, head and neck, endometrial, renal, lung, pancreatic, breast, ovarian, prostate, esophageal, hepatocellular, and gastric carcinomas; osteosarcoma; mesothelioma; melanoma; glioma; medulloblastoma; and rhabdomyosarcoma. They have also been found in T-ALL, Hodgkin lymphoma, anaplastic large-cell non–Hodgkin lymphoma (NHL), AML, B-cell chronic lymphocytic leukemia (CLL), and multiple myeloma. Notch may contribute to carcinogenesis by inhibiting differentiation, inhibiting apoptosis, or promoting proliferation. The intracellular forms of Notch induce transformation when it is expressed with oncoproteins that disable the G1-S checkpoint, such as adenovirus E1A, human papillomavirus E6 and E7, Ras, Myc, or SV40 large T antigen. Notch can activate the expression of several oncogenic pathways via direct or indirect induction of cyclins D1, D3, and A; SKP2; and c-Myc or via activation of PI3K-AKT-mTOR, nuclear factor (NF) κB and NF-κB2, β-catenin, or signal transducer and activator of transcription 3. Notch can also cooperate with oncogenic pathways such as WNT or HER2/Neu.

In addition to its cell-autonomous effects on oncogenic pathways, Notch is involved in tumor-stroma interactions and has a tumor suppressor effect in the epidermis. Notch activity has been reported in cancer stemlike cells (CSCs), which constitute a small subset of cancer cells with a stemlike phenotype that are a reservoir of self-sustaining cells with the ability to self-renew, presumably leading to recurrence. The stemlike phenotype is characterized by enhanced resistance to chemotherapy and radiotherapy, supporting the role of Notch signaling in the maintenance of breast CSCs. Neutralizing monoclonal antibodies directed against Notch 1, 2, and 3 have been used in the clinic. A phase I study of OMP-59R5, a humanized monoclonal antibody that blocks Notch 2 and Notch 3 signaling, was followed by two phase Ib/II trials: the ALPINE trial (Antibody Therapy in First-Line Pancreatic Cancer Investigating Anti-Notch Efficacy and Safety), which is testing OMP-59R5 with gemcitabine and abraxane as first-line therapy in patients with advanced pancreatic cancer, and the PINNACLE trial (Phase Ib/II Investigation of Anti-Notch Antibody Therapy With Cisplatin and Etoposide in Small Cell Lung Carcinoma Efficacy and Safety), which is testing OMP-59R5 combined with cisplatin and etoposide as first-line therapy in patients with extensive-stage small cell lung cancer.

Soluble Dll4-Fc fusion proteins, which bind Notch receptors and prevent their activation by endogenous Dll4, inhibit Notch signaling in endothelial cells, causing disorganized angiogenesis and inhibiting tumor growth. Clinical trials using the OMP-21M18 antibody are ongoing for patients with pancreatic cancer (as first-line therapy with or without abraxane), small cell lung cancer (with carboplatin and pemetrexed), and platinum-resistant ovarian cancer (with paclitaxel).

Nonselective γ-secretase inhibitors (GSIs) are also known as “Notch inhibitors.” The GSI MK-0752 is being investigated in clinical trials in solid tumors and T-ALL. In a phase I trial, MK-0752 was combined with gemcitabine in patients with pancreatic ductal adenocarcinoma.

In a phase Ib trial of the GSI RO4929097 combined with exemestane in metastatic, ER-positive breast cancer, treatment was tolerable, and responses were noted (NCT01149356). The development of RO4929097 has been hampered by its pharmacokinetic liability due to the autoinduction of hepatic metabolism. GSI PF-03084014 is currently in phase I clinical trials for T-ALL and various solid tumors (NCT01981551).

In summary, deregulation of Notch proteins has been associated with cancer development and progression and with the self-propagation of CSCs. Notch-targeted therapies include nonselective GSIs, but various other agents are in development. In HER2-positive breast cancer, the effect of Notch inhibition was recurrence prevention rather than tumor volume decrease.

The fibroblast growth factor receptor (FGFR) family comprises four main members (FGFR1-FGFR4) and encodes membrane tyrosine kinase receptors involved in signaling by interacting with fibroblast growth factors. Activating mutations FGFR3 and FGFR4 exist (^94,95), but amplification of FGFR3 and FGFR4 has been described rarely in cancer. According to the public Cancer Genome Atlas (http://cancergenome.nih.gov), FGFR1 amplification occurred in 3.4% of 10,648 patients and FGFR2 amplification occurred in 0.9% of 8,352 patients. The amplifications were found in lung (16.9%), breast (13.4%), and gastric (5.1%) cancer. Amplification of FGFR1 has been found in lung cancer, SCC of the head and neck, esophageal SCC, and breast and pancreatic cancer. Amplification of FGFR2 has been found in gastric and breast cancer and NSCLC. In a meta-analysis, it was shown that amplification of FGFR1 or FGFR2 may be associated with poor OS in various cancers (^95e). Amplification of FGFR1 was also associated with shorter disease-free survival.

In a phase II study of brivanib (an oral, multitargeted TKI with activity against VEGF and FGFR) in recurrent or persistent endometrial cancer, the ORR was 18.6% (8 of 43 patients; CR, n = 1; PR, n = 7), and 13 patients were progression free at 6 months (⁹⁶). Dovitinib (TKI258) is a TKI that inhibits FGFR, VEGFR, and platelet-derived growth factor receptor (⁹⁷). In a phase II study of dovitinib in metastatic renal cell carcinoma, of 67 patients enrolled, 82.1% were previously treated with one or more VEGFR TKI and one or more mTOR inhibitor. The rates of 8-week overall response and disease control were 1.8% and 52.7%, respectively. The median PFS and the median OS were 3.7 and 11.8 months, respectively (⁹⁷). In a phase II trial of dovitinib in 81 patients with metastatic breast cancer, unconfirmed response or SD for more than 6 months was observed in five (25%) and one (3%) patient(s), respectively, with FGFR1-amplified/HR-positive or FGFR1-nonamplified/HR-positive breast cancer. Other FGFR inhibitors that are being investigated in clinical trials include AZD4547, BAY1187982, BIBF1120, lucitanib (VEGFR-FGFR inhibitor), ponatinib, TAS-120, Debio 1347 (CH5183284), BAY1163877 (pan-FGFR inhibitor), FGF401, BGJ398 (as monotherapy or combined with BYL719), nintedanib, JNJ-42756493, GSK3052230 (combined with paclitaxel and carboplatin or docetaxel or as monotherapy), ARQ 087, BAY1179470, and FPA144. Clinical trials that set FGFR copy number as an inclusion criterion and standardization of FGFR amplification testing may improve these trials (⁹⁵).

Insulin like growth factor (IGF) signaling plays a critical role in the growth and survival of many types of human cancer cells. Although preclinical data of several inhibitors of GF1R were promising in early-phase clinical trials, serious toxicities were observed. Larger randomized phase III trials targeting this pathway failed and led to termination of the anti-IGF1R programs (⁹⁸).

Clinical trials with other inhibitors, such as the murine double minute 2 (MDM2) inhibitor DS-3032b; RO6839921; RO5045337 in combination with doxorubicin in soft tissue sarcoma; RO5503781; HDM201 in combination with LEE011 in liposarcoma; and HDM201 in TP53 wild-type advanced tumors, are ongoing or have been completed. To date, they have demonstrated limited, if any, antitumor activity.

Breakthroughs in technology and the discovery of effective drugs have led to epic advances in the war against cancer. We first demonstrated that, in early-phase clinical trials across tumor types, targeted therapy is associated with high rates of response, PFS, and survival in patients with one targetable molecular alteration (^1,99,100). Next-generation sequencing, circulating DNA and tumor cells, and other profiling will improve our understanding of tumor biology in individual patients. Currently, there is a gap between the plethora of preclinical data and the lack of effective therapies, which is at least partially due to suboptimal drug development for “driver” alterations of human cancer, the high cost of clinical trials and available drugs, and limited access of patients to clinical trials. The complexity of the development of anticancer drugs is evidenced by the high rate of failure of phase III clinical trials of various agents that showed promising results in early-phase studies. Ongoing clinical trials with innovative adaptive study design hold the promise of expediting effective drug development (NCT02152254, NCT01827384, NCT01771458, NCT01248247, NCT01042379, NCT02117167).

Further advances in cancer therapy will be associated with improved technology and bioinformatic analyses to understand dynamic changes in biology and tumor plasticity. Vertical access (changes in time) of cancer biological components to address molecular evolution and horizontal access (changes by site of disease involvement) to address tumor heterogeneity, the interaction between the cancer genome and the epigenome, and the surrounding microenvironment need to be considered. The computational sciences are expected to accelerate drug development by establishing methods to characterize the molecular interactions and analyze a large amount of data about activated pathways, cross talk, and interactions between various components of the cancer intracellular machinery. The power of computational medicine and data sharing is stimulating investigators to develop promising projects that involve the implementation of both bioinformatics and big data analyses. Finally, the development of effective therapeutic strategies, carefully designed clinical trials, and collaborative efforts among key stakeholders in cancer therapy ultimately hold the promise of curing cancer.