Comparison of survivor scores for differentiation therapy of cancer to those for checkpoint inhibition: Half full or half empty

Abstract

Differentiation therapy is directed to the self-renewing cancer stem cells, as well as their progeny transit amplifying cells, to force them to mature to terminal differentiation. Differentiation therapy is effective in treatment of neuroblastomas and myeloid leukemias. Checkpoint inhibition therapy removes blocks to cancer reactive T-killer cells and allows them to react to malignant cells and limit the growth of cancer. The percentage of patients with a given cancer that responds to either therapy is less than hoped for, and the duration of response is variable. Multiplying the response rate (percentage of patients responding to therapy) by the duration of response may be used to derive a survival score for patients treated with differentiation therapy or checkpoint inhibition. By this criterion, differentiation therapy gives better survival scores than checkpoint inhibition. Yet, checkpoint inhibition is considered a great success, mostly because it may be applied to many different types of cancer, and differentiation therapy is considered relatively ineffective because it is limited to a few specific cancers. On the other hand, the cost of checkpoint inhibition treatment is 10–20 times more per patient than that of differentiation therapy. Hopefully, future combined treatments and advances in both approaches will increase the effectiveness of these cancer treatments.

Keywords

Cancer stem cells differentiation therapy checkpoint inhibition survival scores myeloid leukemia

Introduction

Does cancer arise from dedifferentiation of mature cells, by block of differentiation of immature cells (maturation arrest), or both?¹ This has been a long-standing question that is critical for how cancers might be treated. If cancers arise from mature cells, then therapy or prevention could be directed to how to prevent dedifferentiation. On the other hand, if cancer arises from stem or progenitor cells, then therapy or prevention could be directed to removal of the block in differentiation (differentiation therapy (DT)).² Normal tissue renewal involves lineages of cells that mature to terminally differentiated cells so that the number of cells in a normal organ remains essentially the same over time. Cancer arises when there is a block in tissue maturation with accumulation of cells that do not terminally differentiate and continue to proliferate. Pierce et al.³ stated that cancer is a problem of developmental biology; Potter⁴ expressed this as “oncogeny is blocked ontogeny.” DT is directed to removing whatever blocks maturation and allows cancer cells to mature and die.⁵ In practice, DT may be directed both to cancer stem cells (CSCs) and to their progeny.

Teratocarcinoma

The proof of concept for forced differentiation of cancer arises from studies on teratocarcinoma. Pierce and Spears⁶ pointed out that most of the cells of a teratocarcinoma are fully differentiated. The cancer cells are often present in only a small proportion of the tumor mass as “embryoid bodies,” an observation first made by Rudolf Virchow in mid-19th century.⁷ Such observations led others to propose the “embryonal rest” theory of cancer,^8–10 including Cohnheim,¹¹ a student of Virchow. A modern interpretation of the embryonal rest theory is that tissue-determined stem cells are essentially embryonal rests that can not only produce progeny that differentiates into normal mature tissue but also give rise to cancer. Malignant teratocarcinoma stem cells of mice¹² can be converted into normal differentiated tissue cells if transplanted into a normal blastocyst.¹³ In fact, normal-appearing offspring from such blastocysts are chimeras of normal blastocyst cells and transplanted teratocarcinoma cells.^14,15 Just how the normal blastocyst controls the malignant properties of teratocarcinoma stem cells remains poorly understood. In natural germinal cell tumor growth, malignant cells can give rise to benign cells. Embryonal carcinoma is composed of totipotent germinal cells. Differentiation of these cells may produce cells of all three germ layers, including yolk sac and placenta. DT is based on this principle—finding out how to force undifferentiated CSCs to differentiate into non-cancerous cells. This may be accomplished in vitro with retinoic acid (RA, vitamin A). RA reacts with nuclear receptors (RAR-RXR) leading to upregulation of disabled 2 (Dab2), p21, p27, and β-catenin, which inhibits cyclins and proliferation and activates differentiation signaling (Tcf) while decreasing telomerase β (induces senescence) in embryonal carcinoma cells.¹⁶ For many years, it was known that RA could induce differentiation of cultured teratocarcinoma cells. This finding led to attempts to apply DT in clinical trials. However, a recent phase II clinical trial of 16 patients has failed to show significant clinical antitumor activity in patients with chemotherapy-refractory germinal cell tumors.¹⁷ The primary treatment of teratocarcinoma of ovary or testes is surgery combined with chemotherapy or radiation if surgery is not ablative. For testicular tumors, treatment is usually three cycles of BEP (bleomycin, etoposide, and cisplatin) or four cycles of EP (etoposide and cisplatin), essentially cytotoxic therapy. In contrast to expectations, DT has not yet proved to be effective.¹⁸

Neuroblastoma

Neuroblastoma is a cancer arising from neural crest stem cells and is the most common malignancy diagnosed in the first year of life.¹⁹ It is notable that many neuroblastomas discovered in the first year of life spontaneously regress, but in children over 18 months of age, neuroblastomas with unfavorable pathology or MYC-N amplification require intensive treatment, including surgery, radiotherapy, chemotherapy, as well as immunotherapy and DT. It has been known for years that RA treatment reduces proliferation and induces differentiation of neuroblastoma cells in vitro.²⁰ Matthay et al.²¹ reported a 10% response rate with increased survival of 2 years for high-risk neuroblastoma (Figure 1) and later found 29% response rate with increased survival of 2 years in comparison with transplantation of autologous bone marrow alone.²² A recent study on the induction of differentiation of glioma cells in culture²³ concludes,

The existence of a subgroup of patients harboring RA-responsive glioma cells amenable to differentiation therapy and stratifying such patients with a functional test is easily achievable. This provides the first step to reassess the potential of RA in the context of personalized medicine.

Figure 1.

Kaplan–Meier survival curve of high-risk neuroblastoma patients treated with cis-retinoic acid (RA). There is a 10% response rate with a duration of 2 years (delineated by black vertical bars).²¹

Myeloid leukemia

Although DT originally focused on stem cells for teratocarcinoma and epithelial cancers,² DT of myeloid leukemias has proven to be a much better model (Figure 2). In 1996, Druker et al.³¹ identified a small molecular weight compound, identified as SKI571 that competed with the ATP-binding site of the Bcr-Abl tyrosine kinase, blocked the constitutive activation of the leukemic cells, and allowed them to differentiate. This became known as imatinib (Gleevec^®) and could be safely and effectively administered to patients.³² The clinical application of imatinib has increased the 10-year survival of patients with chronic myeloid leukemia (CML) from 20% to up to 91%. In acute promyelocytic leukemia (APL), there is fusion between the promyelocytic leukemia protein and the RA receptor gene, PML/RARα. The PML/RARα fusion product blocks differentiation at the level of the promyelocyte. PML can be treated with all-trans retinoic acid (ATRA), which upregulates ubiquitin activation enzyme-like protein, causing degradation of the gene fusion product and allowing differentiation and apoptosis of the PML cells. This treatment combined with arsenic trioxide has increased 10-year survival of PML patients from 25% to 70% for 48 months (Figure 3).³³

Figure 2.

Stages of maturation arrest, chromosomal translocations, and differentiation treatment of myeloid leukemias. In myeloid leukemia, specific gene translocations are directly responsible for blocking differentiation of cells in the myeloid series. For chronic myeloid leukemia, T9:22, the Philadelphia chromosome²⁴ (Bcr-Abl, break point region-Abelson murine leukemia viral oncogene homolog 1) results in constitutive activation of a tyrosine kinase that keeps myelocytes from differentiation into polymorphonuclear leukocytes.²⁵ In acute promyeloid leukemia, the T15:17, PML/RARα translocation inhibits maturation at the promyelocyte level,²⁶ and in acute myeloid leukemia, various translocations that act to stimulate proliferation or inhibit apoptosis block cells at the hematocytoblast stage.²⁷ Inhibition of STAT3 and FLT3 extends survival in mouse models,^27,28 and a clinical trial of inhibition of FLT3 has also proved beneficial.^29,30

Figure 3.

Kaplan–Meier survival curves for patients with subacute promyeloid leukemia. Treatment with all-trans retinoic acid alone produces a 50% response rate for 48 months when compared to arsenic alone and a 70% response for 86 months when combined with arsenic.³³

In acute myeloid leukemia (AML), there are various translocations that produce fusion products acting on myeloid stem cells (hematocytoblasts) to maintain stemness. For AML, there appears to be two types of mutations: Class I, which increases proliferation (FLT3, KIT), and Class II, which blocks maturation. So far, the major therapeutic approach in treatment of AML is various doses and schedules of delivery of cytotoxic drugs, such as cytarabine and anthracyclines, but there are continued efforts to improve this toxic and relatively ineffective approach. Potential new targets for therapy are directed toward inhibition of interleukin (IL)-3R activation in a subset of patients with mutant FLT3,^34,35 blocking STAT3 (epidermal growth factor receptor (EGFR) and IL-6) signaling pathways³⁶ and targeting MUC-1 to induce differentiation and increase survival.³⁷ It remains to be seen whether a subset of AML patients may benefit as much as other myeloid leukemia patients. It is quite likely that, as in the case of AML, identification of maturation arrest mutations or translocations in other cancers will be identified and appropriate blocking agents and effective treatments developed, for example, blocking pathways known to maintain stemness, such as Wnt, Oct-4, Notch, BMP, JAK, and others.³⁸

Comparison of DT and checkpoint inhibition

Antibodies to checkpoint inhibitors block the action of CTLA4^39,40 and PD-1^41–43 produced by CD3+ T-cytotoxic cells as off signals to prevent cancer-reactive T cells from proliferating and inhibiting tumor immunity. For example, PD-1 inhibitors have enabled patients with untreatable tumors, such as melanoma and lung cancer, to survive for years, but this effect is not seen in most patients. A recent review of 23 trials of treatments targeted to PD-1 or PDL1 show a response rate from 5% for sarcoma to over 80% in Hodgkin’s disease with a response around 20% for most cancers, including non-small-cell lung cancer and melanoma.⁴⁴ In addition to the percent response rate, a critical factor is the duration of response. By multiplying the percent response by the duration, the survival score can be obtained.

Kaplan–Meier curves and survival scores

The analysis of results of clinical trials depends on Kaplan–Meier survival curves.⁴⁵ In the study by Herbst et al.⁴⁶ shown in Figure 4, the percentage of patients treated with anti-PD-1 alive during the time of study is compared to the percent remaining that is treated conventionally with docetaxel. The top curve (A) shows the results for patients whose cancer cells are greater than 50% PD-1+ by immunofluorescence; the lower curve (B) shows the results for all patients. The most complicated part of this type of analysis is to account for the patients who drop out of the study and are lost to follow-up.⁴⁵ Such patients are removed from the study (censured). The small vertical lines on the curves in Figure 4 denote censured patients. The percent response is determined by the number of patients left in the study when the number of patients has become too low to be meaningful. In the study shown, this occurs at 15 months. The duration of the response is when there is a clear separation of the curves. Thus, in Figure 4, the response rate for A is 30% and the duration is 10 months. The survival score is obtained by multiplying the percent response rate by the duration of response. Thus, in A, this is 30 × 10 = 300. For curve B, the response rate is 20% and the duration is 8 months, for a survival score of 160 (20 × 8). A caveat for the survival score is that it gives equal weight to percent patient response and duration of response. This is an arbitrary choice that may not be agreed upon by everyone.

Figure 4.

Kaplan–Meier survival curves for anti-PD-1 therapy of non-small-cell lung cancer. (A) Patients with >50% PD-1+ cells and (B) all patients.⁴⁶

Comparative survival scores

Some representative survival scores are listed in Table 1. There are three levels for DT: very high for leukemias (2000–6000), moderate for neuroblastoma (240–480), and low for non-small-cell lung carcinoma (NSCLC) (survival score 64). Checkpoint inhibition (CPI) therapy has similar but somewhat lower levels: high for Hodgkin’s lymphoma (1145);⁴⁷ moderate for melanoma,⁴⁸ renal cell carcinoma,^43,49 Merkel-cell carcinoma,⁵⁰ and relapsed hematologic cancers (180–512);⁵¹ and variable or low for NSCLC (80–420).^52–54 Statistical comparison of the combined treatment response for DT versus CPI was p = 0.0114 by non-paired t-test. The lower than expected scores for CPI treatment of NSCLCs is almost certainly related to the fact that most studies have been done on previously treated patients. This comparison is compromised by the attempt to compare effects of different treatments for different diseases with different natural histories. However, this approach does allow a comparison of overall treatment outcomes.

Table 1.

Representative survival scores for patients treated with DT compared to CPI therapy.

Cancer type	% Response	Duration (months)	Survival score
DT
Neuroblastoma	10	24	240
Neuroblastoma	29	24	696
CML	20	96	1920
CML	5	120	600
APL	70	96	2400
APL	50	36	1800
APL	35	36	1260
AML	10	66	660
Combined differentiation and chemotherapy
APL (chemotherapy)	20	48	960
APL (chemotherapy + ATRA)	70	48	6020
CPI
Melanoma	28	1.9–24	420
NSCLC	16	6	96
NSCLC	30	10	300
NSCLC	18	1.9–30	180
NSCLC	20	4	80
Renal cell carcinoma	27	5.6–22.3	300
Hodgkin lymphoma	69	16.6	1145
Merkel-cell carcinoma	56	2.2–9.7	336
Relapsed hematologic cancers	32	16	512
Combined CPI and chemotherapy
NSCLC (chemotherapy)	29	8.9	258
NSCLC (chemotherapy + CPI	55	19	1045
Combined CPI and radiotherapy
NSCLC (radiotherapy)	27	5.6	151
NSCLC (radiotherapy+ CPI)	44.2	16.8	743

DT: differentiation therapy; CPI: checkpoint inhibition; CML: chronic myelogenous leukemia; AML: acute myeloid leukemia; ATRA: all-trans retinoic acid; NSCLC: non-small-cell lung carcinoma; APL: acute promyelocytic leukemia.

Improving checkpoint inhibitor treatment

Patient selection

Prediction of which patients will respond to checkpoint inhibitors is critical to selection of therapy.⁵⁵ First, for patients with cancers who express PD-L1, the receptor for PD-1 will respond to PD-1 blocking therapy better than those who do not,⁵⁶ but PD-L1 expression does not always accurately predict which patients will respond to CPI. Second, tumors with CD8+ T-cell infiltrate will respond better than tumors that do not. It is theorized that tumors that have low CD8+ T cells require more aggressive therapy, such as combined anti-PD-1/ anti-CTLA-4 treatments (see below). In addition, since transforming growth factor (TGF)-β inhibits T cells from entering tissues, combined treatment with CPI and the TGF-β blocker, M7824, is being tested. Third, another marker is the tumor mutational burden (TMB). Since each mutation in a tumor may generate a neoantigen, the more mutations that a tumor has, the more likely it is to respond to CPI. Fourth, the gut microbiome may affect CPI outcome. An ongoing study of the effect of an oral treatment with a mix of microbes that match that from anti-PD-1 responders will test this hypothesis.

Other cancer targets (NKG2D and IL-1R8)

The cell surface molecule, natural killer group 2D (NKG2D), is highly expressed on natural killer (NK) cells and T cells. NKG2D reacts with a major histocompatibility complex (MHC)-like class I ligand, which is upregulated in cells under stress.⁵⁷ This ligand–receptor interaction is a major pathway for the detection and elimination of damaged, infected, or transformed cells.⁵⁸ The NKG2D ligand is highly expressed on cancer cells and virus-infected cells, but not widely on normal tissue cells,⁵⁹ so this reaction has potential as a new target for immunotherapy.⁵⁷ Expression of NKG2D ligands on the surface of cancer cells correlates with a good prognosis, whereas soluble NKG2D ligand indicates a poor prognosis and negatively impacts anti-PD-1 CPI therapy. Both chemotherapy and radiation increase the expression of NKG2D ligand on cancer cells and enhance tumor immunity. Monoclonal antibodies to NKG2D increase presentation of tumor antigens to activate T cells as well as promote antibody-dependent cellular cytotoxicity of tumor cells by NK cells, neutrophils, and macrophages through Fc receptors.⁵⁹ A variety of agents, most prominent being the histone deacetylase (HDAC) inhibitor valproic acid, increase membrane NKG2D expression, whereas others can decrease secretion of non-membrane NKG2D. These have been proposed as potential treatments for cancer.^57,60

Interleukin-1 receptor family member R8 (IL-1R8), through binding to IL-37, negatively regulates interleukin receptor and toll-like receptor signaling and inhibits both innate and adaptive immunity.⁶¹ IL-8R binding and activation are active in almost all inflammatory and infectious reactions where it serves to limit inflammation.⁶¹ Colon tumors express less IL-8R than normal colonic mucosa cells and may thus be better able to escape tumor-reactive cells. The possible effect of IL-1R8 in resistance to CPI or as a target for therapy remains to be determined.

Combination therapy

The effect of combination therapy for both DT and CPI is shown in Table 1. Combined ATRA DT and arsenic trioxide improves the survivor score by sixfold over arsenic trioxide alone. There are approximately 2000 clinical trials now underway using six Food and Drug Administration (FDA)-approved humanized monoclonal antibodies to CTLA-4, PD-1, and PL-L1⁶² for cancer treatment for various cancers including melanoma, multiple myeloma, leukemia, lymphoma, glioblastoma, gastric, renal cell, bladder, colorectal, hepatocellular, breast, non-small-cell lung, prostate, and other cancers.^62,63 However, as indicated above, about 60%–70% of patients do not respond to single-agent CPI therapy (Table 1). This result has stimulated the use of checkpoint inhibitors in combination with conventional treatments,⁶⁴ such as cisplatin, 5-fluorourcil, and lenalidomide,⁶⁵ as well as targeted therapy and radiation therapy.⁶⁵ Chemotherapy may increase tumor antigenicity, disrupt immune suppression, enhance T-cell responses, or increase antigenicity or major histocompatibility marker expression. FDA approval of combinations of pembrolizumab (anti-PD-1) and chemotherapy has resulted in multiple new clinical trials.⁶⁶ The early results of a combination trial of pembrolizumab (anti-PD-1) and carboplatin and pemetrexed for NSCLC patients resulted in 55% response rate for 13 months (survival score 715) compared to 29% response rate for 6 months (survival score 132) for pemetrexed alone⁶⁷ with updated survival time of 19 months (survival score 1045) for combination compared to 8.9 months (survival score 258) for pemetrexed.⁶⁸ In a phase III trial for squamous NSCLC, combining pembrolizumab with taxene and carboplatin resulted in a survival score of 928 for combined therapy versus 168 for chemotherapy.⁶⁹ Similar results have been found in other clinical trials.⁶⁶ However, the combined chemo-immunotherapy approach is just beginning and much needs to be done to determine the optimal dose/scheduling program for therapy. Table 1 indicates that the survival score for combined therapy (1045) is much better than chemotherapy alone (258) or CPI alone (80–300).

Molecular targeted therapy (signaling pathways) and CPI therapy

In melanoma, BRAF mutations are associated with decreased antitumor immunity, elevation of PD-L1 and accumulation of regulatory T cells, and downregulation of MHC-1,⁶⁹ which can at least temporally be reversed by BRAF inhibitors. The combination of checkpoint inhibitors, ipilimumab (anti-CLTA-4) and nivolumab (anti-PD-1), with kinase (MAPK) inhibitors and inhibitor of BRAF mutation resulted in tumor growth inhibition,⁷⁰ but other combined studies have been complicated by severe liver and gastrointestinal toxicities. Much more work is needed to determine whether this approach is beneficial. Vascular endothelial growth factor (VEGF) stimulates tumor vascularization and is immunosuppressive. Multiple clinical trials are currently underway in breast, gynecological, lung, and other cancers, using bevacizumab and pembrolizumab, inhibitors of VEGF.⁷¹ These show some promising findings, such as an increase in intra-tumoral T cells, but definitive positive therapeutic effects have not yet been convincingly demonstrated.⁶⁶ An additional target for combination therapy is CD-40.^72,73 CD-40 is a co-stimulatory signal that binds to CD-40L and enhances the immune response to tumors. Anti-CD40 is an agonist for CD-40 that activates this pathway and is being tested in combination with nivolumab (anti-PD-1).

Radiotherapy

Radiotherapy induces cell death and can stimulate antitumor immunity. A clinical trial of chemoradiation followed by durvalumab (anti-PD-L1) in stage III NSCLC resulted in a response rate of 44.2% for 16.8 months (survival score 743) compared to 27% for 5.6 months in those receiving chemoradiation and placebo (survival score 151) or anti-PD-1 alone (80–300), an encouraging early result.⁷⁴

Adverse effects

From 5% to 87% of patients treated with checkpoint blockade experience grade 3–4 treatment–related adverse effects with up to 100% having less serious effects.^44,75 Serious effects include pneumonitis, hepatitis, colitis, endocrinopathy, and nephritis that may require cessation of treatment. In contrast, with RA treatment of APL or neuroblastoma, only mild dryness of the lips and skin with occasional headaches and digestive symptoms have been reported.^76,77

Cost

The estimated cost per cancer patient per year for CPI therapy is US$1,000,000,⁷⁸ compared to US$150,000 for Gleevec⁷⁹ and US$60,000 for RA plus arsenic trioxide for myeloid leukemia.⁸⁰

Conclusion

Based on survivor scores, DT is somewhat better than CPI therapy. However, CPI can be applied to many different types of cancer, whereas DT is limited to neuroblastoma and myeloid leukemias. There should be continued search for specific agents for DT, as well as the possibility of combining DT with CPI. The limiting factor is the relatively small number of cancer types that can be successfully treated with DT. On the other hand, CPI treatment is roughly 10–20 times more expensive than DT and has much more severe adverse effects.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Stewart Sell

Zoran Ilic

References

Liu

The dualistic origin of human tumors. Semin Cancer Biol 2018; 53: 1–16.

Sell

Pierce

GB.

Maturation arrest of stem cell differentiation is a common pathway for the cellular origin of teratocarcinoma and epithelial cancers. Lab Invest 1994; 70: 6–22.

Pierce

Shikes

Fink

LM.

Cancer: a problem of developmental biology. Englewood Cliffs, NJ: Prentice Hall Inc., 1978, pp. 1–242.

Potter

VR.

Phenotypic diversity in experimental hepatomas: the concept of partially blocked ontogeny. Br J Cancer 1978; 38(1): 1–23.

Sell

Stem cell origin of cancer and differentiation therapy. Crit Rev Oncol Hematol 2014; 51: 1–28.

Pierce

Spears

WC.

Tumors as caricatures of the process of tissue renewal: prospects for therapy by directing differentiation. Cancer Res 1998; 48: 1196–1204.

Virchow

Die krankhafter Geschwülste. Berlin: Hirschwald, 1863.

Récamier

JCA

. Recherches sur le traitement du cancer par la compression méthodique simple ou combinée, et sur l’histoire général de la même maladie. Paris: Gabon, 1829.

Remark

Ein Beitrag zur Entwicklungsgeschichte der krebshaften Geschwülste. Dtsch Klin 1854; 16: 170–174.

10.

Durante

Nesso fisio-pathologico tra la struttura dei nei materni e la genesi di alcuni tumori maligni. Arch Memor Observ Chir Pract 1874; 11: 217–226.

11.

Cohnheim

Congenitales, quergestreiftes Muskelsarkon der Nireren. Virchows Arch 1875; 65: 64–69.

12.

Stevens

LC.

Experimental production of testicular teratomas in the mouse. Int J Androl 1981; 4(s4): 54–59.

13.

Brinster

RL.

Effect of cells transferred into the mouse blastocyst on subsequent development. J Exp Med 1974; 140(4): 1049–1056.

14.

Mintz

Illmensee

Normal genetically mosaic mice produced from malignant teratocarcinoma cells. Proc Natl Acad Sci USA 1975; 72(9): 3585–3589.

15.

Papaioannou

McBurney

Gardner

, et al. Fate of teratocarcinoma cells injected into early mouse embryos. Nature 1975; 258(5530): 70–73.

16.

Chambon

A decade of molecular biology of retinoic acid receptors. FASEB J 1966; 10: 940–945.

17.

Moasser

Motzer

Khoo

K-S

, et al. All-trans retinoic acid for treating germ cell tumors: in vitro activity and results of a phase II trial. Cancer 1995; 76(4): 680–686.

18.

Cancer.Net Editorial Board. Testicular cancer: treatment options, https://www.cancer.net/cancer-types/testicular-cancer/treatment-options (2018, accessed 17 December 2018).

19.

Matthay

Maris

Schleiermacher

, et al. Neuroblastoma. Nat Rev Dis Primers 2016; 2: 16078.

20.

Tsokos

Scarpa

Ross

, et al. Differentiation of human neuroblastoma recapitulates neural crest development. Study of morphology, neurotransmitter enzymes, and extracellular matrix proteins. Am J Pathol 1967; 128: 484–496.

21.

Matthay

Reynolds

Seeger

, et al. Long-term results for children with high-risk neuroblastoma treated on a randomized trial of myeloablative therapy followed by 13-cis-retinoic acid: a Children’s Oncology Group Study. J Clin Oncol 2009; 27: 1007–1012.

22.

Matthay

Villablanca

Seeger

, et al. Treatment of high-risk neuroblastoma with intensive chemotherapy, radiotherapy, autologous bone marrow transplantation and 13-cis-retinoic acid. Children’s Cancer Group. N Engl J Med 1999; 341(16): 1165–1173.

23.

Dreyfus

El-Atifi

Court

, et al. Reprogramming glioma cell cultures with retinoic acid: additional arguments for reappraising the potential of retinoic acid in the context of personalized glioma therapy. Glioma 2018; 1: 66–78.

24.

Nowell

PC.

Diagnostic and prognostic value of chromosome studies in cancer. Ann Clin Lab Sci 1974; 4(4): 234–240.

25.

Deininger

Vieira

Mendiola

, et al. BCR-ABL tyrosine kinase activity regulates the expression of multiple genes implicated in the pathogenesis of chronic myeloid leukemia. Cancer Res 2000; 60(7): 2049–2055.

26.

Longo

Pandolfi

Biondi

, et al. Rearrangements and aberrant expression of the retinoic acid receptor alpha gene ns acute promyelocytic leukemias. J Exp Med 1990; 172: 1571–1574.

27.

Shen

Zhu

Fan

, et al. Gene mutation patterns and their prognostic impact in a cohort of 1185 patients with acute myeloid leukemia. Blood 2011; 118(20): 5593–5603.

28.

Jiang

Ferchen

, et al. Targeted inhibition of STAT/TET1 axis as a therapeutic strategy for acute myeloid leukemia. Nat Commun 2017; 13: 2099.

29.

Kelly

Boulton

, et al. CT53518, a novel selective FLT3 antagonist for the treatment of acute myelogenous leukemia (AML). Cancer Cell 2002; 1(5): 421–432.

30.

Stone

Mandrekar

Sanford

, et al. Midostaurin plus chemotherapy for acute myeloid leukemia with a FLT3 mutation. N Engl J Med 2017; 377: 454–464.

31.

Druker

Tamura

Buchdunger

, et al. Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nat Med 1996; 2(5): 561–566.

32.

Druker

Talpaz

Resta

, et al. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 2001; 344(14): 1031–1037.

33.

Kantarjian

Acute myeloid leukemia—major progress over four decades and glimpses into the future. Am J Hematol 2016; 91(1): 131–145.

34.

Chalandon

Schwaller

Targeting mutated protein tyrosine kinases and their signaling pathways in hematologic malignancies. Haematologica 2005; 90(7): 949–968.

35.

Stone

RM.

Which new agents will be incorporated into frontline therapy in acute myeloid leukemia. Best Pract Res Clin Haematol 2017; 30(4): 312–316.

36.

Redell

Ruiz

Alonzo

, et al. Stat3 signaling in acute myeloid leukemia: ligand-dependent and -independent activation and induction of apoptosis by a novel small-molecule Stat3 inhibitor. Blood 2011; 117(21): 5701–5709.

37.

Stroopinsky

Rajabi

Nahas

, et al. MUC1-C drives myeloid leukaemogenesis and resistance to treatment by a survivin-mediated mechanism. J Cell Mol Med 2018; 22(8): 3887–3898.

38.

Ivanova

Dobrin

, et al. Dissecting self-renewal in stem cells with RNA interference. Nature 2006; 442(7102): 533–538.

39.

Brunet

Denizot

Luciani

, et al. A new member of the immunoglobulin superfamily—CTLA-4. Nature 1987; 328: 267–270.

40.

Leach

Krummel

Allison

JP.

Enhancement of antitumor immunity by CTLA-4 blockade. Science 1996; 271(5256): 1734–1736.

41.

Agata Kawasaki

Nishimura

Ishida

, et al. Expression of the PD-1 antigen on the surface of stimulated mouse T and B lymphocytes. Int Immunol 1996; 8(5): 765–772.

42.

Francisco

Sage

Sharpe

AH.

The PD-1 pathway in tolerance and autoimmunity. Immunol Rev 2010; 236: 219–242.

43.

Topalian

Hodi

Brahmer

, et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med 2012; 366: 2443–2454.

44.

Borcherding

Kolb

Gullicksrud

, et al. Keeping tumors in check: a mechanistic review of clinical response and resistance to immune checkpoint blockade in cancer. J Mol Biol 2018; 430(14): 2014–2029.

45.

Rich

Neely

Paniello

, et al. A practical guide to understanding Kaplan-Meier curves. Otolaryngol Head Neck Surg 2010; 143(3): 331–336.

46.

Herbst

Baas

Kim

, et al. Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small cell lung cancer (KEYNOTE-010): a randomised controlled trial. Lancet 2016; 387: 1540–1550.

47.

Armand

Engert

Younes

, et al. Nivolumab for relapsed/refractory classic Hodgkin lymphoma after failure of autologous hematopoietic cell transplantation: extended follow-up of the Multicohort Single-Arm Phase II CheckMate 205 trial. J Clin Oncol 2018; 36(14): 1428–1439.

48.

Robert

Schachter

Long

, et al. Pembrolizumab versus ipilimumab in advanced melanoma. N Engl J Med 2015; 372: 2521–2532.

49.

Motzer

Escudier

McDermott

, et al. Nivolumab versus everolimus in advanced renal-cell carcinoma. N Engl J Med 2015; 373: 1430–1347.

50.

Nghiem

Bhatia

Lipson

, et al. PD-1 blockade with pembrolizumab in advanced Merkel-cell carcinoma. N Engl J Med 2016; 374(26): 2542–2552.

51.

Davids

Kim

Bachireddy

, et al. Ipilimumab for patients with relapse after allogeneic transplantation. N Engl J Med 2016; 375: 143–153.

52.

Clandos

Remarkable immunotherapy trial changes advanced lung cancer treatment, https://labblog.uofmhealth.org/lab-report/remarkable-immunotherapy-trial-changes-advanced-lung-cancer-treatment (2018, accessed 17 December 2018).

53.

Carbone

Reck

Paz-Ares

, et al. First-line nivolumab in stage IV or recurrent non-small-cell lung cancer. N Engl J Med 2017; 376: 2415–2426.

54.

Rittmeyer

Barlesi

Waterkamp

, et al. Atezolizumab versus docetaxel in patients with previously treated non-small-cell lung cancer (OAK): a phase 3, open-label, multicentre randomised controlled trial. Lancet 2017; 389: 255–265.

55.

Friedrich

MJ.

Immunotherapy 2.0: improving the response to checkpoint inhibitors. JAMA 2019; 321(2): 131–133.

56.

Herbst

Soria

J-C

Kowanetz

, et al. Predictive correlates of response to the anti-PD-I1 antibody MPDL3289A in cancer patients. Nature 2014; 515: 563–567.

57.

Schmiedel

Mandelboim

NKG2D ligands-critical targets for cancer immune escape and therapy. Front Immunol 2018; 9: 2040.

58.

Zingoni

Molfetta

Fionda

, et al. NKG2D and its ligands: “one for all, all for one.” Front Immunol 2018; 9: 476.

59.

Spear

M-R

Sentman

M-L

, et al. NKG2D ligands as therapeutic targets. Cancer Immun 2013; 13: 8.

60.

Cathcart

Pulkoski-Gross

Cao

Targeting matrix metalloproteinases in cancer: bringing new life to old ideas. Genes Dis 2015; 2(1): 26–34.

61.

Molgora

Barajon

Mantovani

, et al. Regulatory role of IL-1R8 in immunity and disease. Front Immunol 2016; 7: 149.

62.

Hargadon

Johnson

Williams

CJ.

Immune checkpoint blockade therapy for cancer: an overview of FDA-approved immune checkpoint inhibitors. Int Immunopharmacol 2018; 62: 29–39.

63.

Alsaab

Alzhrani

, et al. PD-1 and PD-L1 checkpoint signaling inhibition for cancer immunotherapy: mechanism, combinations, and clinical outcome. Front Pharmacol 2017; 8: 561.

64.

Ledford

Cocktails for cancer with a measure of immunotherapy. Nature 2016; 532(7598): 162–164.

65.

Morrissey

Yuraszeck

, et al. Immunotherapy and novel combinations in oncology: current landscape, challenges, and opportunities. Clin Transl Sci 2016; 9(2): 89–104.

66.

Yan

Kumar

Finnes

, et al. Combining immune checkpoint inhibitors with conventional cancer therapy. Front Immunol 2018; 9: 1739.

67.

Langer

Gadgeel

Borghaei

, et al. Carboplatin and pemetrexed with or without pembrolizumab for advanced, non-squamous non-small cell lung cancer: a randomized phase 2-cohort of the open-label KEYNOTE-021 study. Lancet Oncol 2016; 17: 1497–1508.

68.

Borghaei

Langer

Gadgeel

, et al. 24-month overall survival from KEYNOTE-021 cohort G: pemetrexed and carboplatin with or without pembrolizumab as first-line therapy for advanced nonsquamous non-small cell lung cancer. J Thorac Oncol 2019; 14(1): 124–129.

69.

Paz-Ares

Luft

Tafreshi

, et al. Phase 3 study of carboplatin-paclitaxel/nab-paclitaxel (chemo) with or without pembrolizumab (pembro) for patients (Pts) with metastatic squamous (Sg) non-small cell lung cancer (NSCLC). J Clin Oncol 2018; 36(15 Suppl.): 105.

70.

Azijli

Stelloo

Peters

, et al. New developments in the treatment of metastatic melanoma: immune checkpoint inhibitors and targeted therapies. Anticancer Res 2014; 34(4): 1493–1505.

71.

McCabe

Creasy

CL.

EZH2 as a potential target in cancer therapy. Epigenomics 2014; 6(3): 341–351.

72.

Elgueta

Benson

de Vries

, et al. Molecular mechanism and function of CD40/CD40L engagement in the immune system. Immunol Rev 2009; 229(1): 152–172.

73.

Zhang

Chen

, et al. The CD40/40L system: a new therapeutic target for disease. Immunol Lett 2013; 153: 58–61.

74.

Antonia

Villegas

Daniel

, et al. Durvalumab after chemoradiotherapy in stage III non-small-cell-lung cancer. N Engl J Med 2017; 377: 1919–1929.

75.

Michot

Bigenwald

Champiat

, et al. Immune-related adverse events with immune checkpoint blockade: a comprehensive review. Eur J Cancer 2016; 54: 139–148.

76.

Huang

Chen

, et al. Use of all-transretinoic acid in the treatment of acute leukemia. Blood 1988; 72: 567–572.

77.

Dulaney

Murgatroyd

RJ.

Use of trans-retinoic acid in the treatment of acute promyelocytic leukemia. Ann Pharmacother 1993; 27: 211–214.

78.

Andrews

Treating with checkpoint inhibitors—figure $1 million per patient. Am Health Drug Benefits 2015; 8(Spec Issue): 9.

79.

Kantarjian

The arrival of generic imatinib into the U.S. market: an educational event. The ASCO post, http://www.ascopost.com/issues/may-25-2016/the-arrival-of-generic-imatinib-into-the-us-market-an-educational-event/ (2016, accessed 10 March 2019).

80.

Tallman

Lo-Coco

Barnes

, et al. Cost-effectiveness analysis of treating acute promyelocytic leukemia patients with arsenic trioxide and retinoic acid in the United States. Clin Lymphoma Myeloma Leuk 2015; 15(12): 771–777.