Dr. Kittles discusses the race-based health disparities in prostate cancer, and one genetic-environmental crossover that is clinically significant.

There are significant race-based disparities in prostate cancer care and outcomes. Recent research has shed more light on the divergent causes of these disparities. 

 spoke with Rick Kittles, Professor and Founding Director of the Division of Health Equities within the Department of Population Sciences at City of Hope Comprehensive Cancer Center in Duarte, California, where he is also Associate Director of Health Equities. Among other health disparity topics, Dr. Kittles conducts research on why African-Americans have a disproportionately higher burden of prostate cancer compared to European-Americans.

 First, could you lay out the research questions that you address, as they relate to health disparities and prostate cancer specifically among African-Americans? What are the main issues?

 There are several issues dealing with prostate cancer disparities that we focus on. What we are attempting to do is to try to understand why there’s an increased burden and incidence of prostate cancer for men of West African descent in particular. There are mortality differences, too, and other different outcomes compared with other populations. We’re trying to understand if there is a strong genetic component to this, or is it environmental or the interaction of both of these factors? We’re doing a lot of genetic studies and behavioral and dietary studies in this area.

 You and your colleagues recently conducted a study analyzing the role of calcium and vitamin D in prostate cancer pathogenesis and progression. Could you tell us about the genesis of this study and what you found?

 The calcium study emerged after we started studying vitamin D. Vitamin D, calcium, and parathyroid hormone interact and regulate each other in one way or another, and this relates not only to what we historically have known in terms of bone mineralization and bone development, but also to immune surveillance. This is something that we started to really focus in on-specifically the role vitamin D plays in immune surveillance. Our calcium study showed that there was this higher incidence or higher risk of aggressive prostate cancer among African-American men whose daily calcium intake was over 800 mg.[1] This was quite surprising, but there have been other studies that have shown similar results.

 Can you talk about vitamin D research and how vitamin D might be linked to prostate cancer?

 Is vitamin D linked to prostate cancer? Some have said that there’s an association and others say there is not. A lot of the differences in results has to do with how you are collecting and analyzing data. Our group has found that there are many different modifiers of serum vitamin D levels, and there are differences in levels of prostatic vitamin D, the amount of vitamin D in the prostate. And those levels, whether they are sufficient or deficient, could lead to differences in gene expression, many of which are related to immune response, inflammation, and immune surveillance. Vitamin D deficiency can lead to altered immune surveillance in the prostate tissue, which could potentially lead to prostate cancer.

 Are there other factors that may affect this higher prevalence of prostate cancer among African-Americans that you have found in your research?

 When we look at the difference between African-American men and men of European descent, we find that genes are playing a strong role. In fact, upwards of maybe 40% to 50% of the difference we see is genetic. There are some strong genetic signals that emerged over the last 10 years related to a region on chromosome 8 that appear to be enriched in men of West African descent. (The region of risk on chromosome 8, specifically, is the 8q24 region.) There may be some sort of increased inherited predisposition based on ancestry that could be driving a lot of the differences that we see. We don’t know the mechanism of action, but we do know that the markers and biomarkers are pretty consistent and replicable.

 Using those genetic markers, has your group or other groups followed up this work to potentially screen men for higher risk of prostate cancer or does this still need to be confirmed in additional studies?

 Well, a lot of this work on biomarkers was work from a consortium, what we call GWAS or genome-wide association studies on prostate cancer where multiple groups around the country put their samples together for these big meta-analyses. This was not just our group; many groups actually helped find the signals. The sad part is that there hasn’t been a focus, let’s say from the National Institutes of Health or other funding agencies, to really say, “Look, now there are these signals that have been replicated across the world, actually.” We see these signals in all populations: European, Asian, Indian, South African. The problem is that there has not been a concerted effort to say, “Let’s look at mechanism of action now to see the real inherited mechanism that is increasing this burden.” I say this all the time: once you find something and there’s next steps that could be done, you need to have that increased attention on those next steps if you really want to peel apart what’s going on. Right now, there are several groups still working on this––not as many as before––and there’s also a marker panel that is just emerging and might actually be able to predict risk for certain outcomes. But still, no clear evidence of mechanism has been shown yet.

 Are you conducting other studies to identify other potential factors or the role of the environment in this health disparity connection?

 The vitamin D study was actually a gene environment study. We’re looking at genes, their pathways, the steroid synthesis pathway, and the steroid signaling pathways. We are also looking at dietary behavioral factors that influence vitamin D levels in serum and in the prostate. This is a really good model of a disparity project that has a strong gene-environment interaction. We are also studying methylation and dietary folate, folic acid in particular. We are looking at some of the clinical outcomes as they relate to the heterogeneity among these tumors. What we’re finding is that there’s no overburden of particular genetic changes in these prostate tumors. There’s a subset of different types of signals and mutations in various tumors in the African-American men who have prostate cancer. What we’re trying to do now is to uncover some of the outcomes from these particular signatures. For example, HER2, which is a clear breast cancer subtype and is used as a biomarker in breast cancer, actually has a therapeutic potential. We need to identify markers like this for prostate cancer.

 More recently, you had a study that looks at different cancer cell lines and highlighted disadvantages for minorities in accessing personalized therapy strategies for prostate and other tumor types. Can you tell us about that work and what you found?

 The challenges and problems have been the lack of diversity in biorepositories and biospecimens. This has limited the progress in disparities research. What we wanted to do was really examine if, in fact, cell lines that were labeled or characterized as African-American were really from African-American patients. And, in fact, many were not,[2] and so this is hopefully bringing some attention to this fact that we still, in particular for prostate cancer, have this enormous disparity difference among ethnic groups. We really need to focus in on collecting more diverse samples for our research models of prostate cancer, and also other tumor types.

 Lastly, can you talk about your efforts and efforts by the broader scientific community to make bioscience research more inclusive to different ethnic populations?

 My whole career has been focused on studying disparities in Black and Hispanic populations. I focus on going out, educating the community, getting them excited about research, and then getting them to participate. For me, there is no challenge in terms of getting access to samples, of getting people excited about the research. I think a lot of times the community really just needs to see themselves in the research. When I say that I mean they need to see something that is impactful for them, something that they want to see resolved in their community. They also want to see themselves reflected in the research, as it relates to participating researchers, clinicians, nurses, and students. Let’s say the principal investigator of a study is not of African-American descent and staff isn’t either. This does play a role in the ability of the researcher to recruit a diverse population and get diverse samples.

 Anything else you would like to add on your research and increasing diversity in scientific research?

 I think this issue of increasing diversity in scientific research is going to have a major role in the future of research here in the United States. As everybody knows, the country is becoming more and more diverse. If we do not have adequate representation, not just in biomedical samples and biospecimen repositories, but also in representation in the pipeline and workforce, that lack of diversity is going to have some major ramifications downstream. I get excited when I go out into the community and talk about these issues. I’m trying to get folks to be more aggressive in terms of their advocacy for themselves and support for diversity in the biomedical workforce.

Dr. Kittles has no significant financial interest in or other relationship with the manufacturer of any product or provider of any service mentioned in this article.

Unraveling the Research on Prostate Disparities

Although we have known for decades that African American men with prostate cancer are at an extreme disadvantage, scientists are still seeking the translational answers that might reverse this reality. The literature includes a wide variety of genetic, social, and environmental theories that they hope will explain, individually or in combination, the higher risk and poorer outcomes seen in these patients. Dr. Kittles has discussed some of these in his interview, and focuses specifically on a genetic predisposition and vitamin D. This and other factors have been examined, including diet, exercise, testosterone levels, androgen receptor variations, immunological responses, and many more. A major component includes access to medical care and treatment. In some equal access situations, such as the military, disparities in survival seemed to be lessened and in some cases eliminated. We published an article in 2004 looking at outcomes of African-American men in the Henry Ford system regarding local prostate cancer outcomes. Some of the discrepancies appear to be related to choices of therapy. When African American men, who represented 30% of the men in the system, chose the same treatment as Caucasians, they had similar survival. [1] Interestingly, Dr Oliver Sartor, reported recently that African-American men with advanced prostate actually had a survival advantage over Caucasian man when treated with immunotherapeutic regimen of sipuleucel-T (Provenge). As we continue to explore all the research on prevalence and prognosis, we may find that we can increase favorable outcomes based on multiple factors. I applaud Dr. Kittles‘s work and appreciate the comments.

Dr. Crawford has no significant financial interest in or other relationship with the manufacturer of any product or provider of any service mentioned in this article.

[1] Tewari A, Johnson CC, Divine G, Crawford ED, Gamito EJ, Demers R and Menon: Long-term survival probability in men with clinically localized prostate cancer: a case-control, propensity modeling study stratified by race, age, treatment and comorbidities. Jour Urology 171(4):1513–1519, 2004.

[2] Sartor AO, Andrew J, Armstrong, CA, et al.

 Batai K, Murphy AB, Ruden M, et al. Race and BMI modify associations of calcium and vitamin D intake with prostate cancer. BMC Cancer. 2017; 17(1):64.

 Hooker SE, Woods-Burnham L, Bathina M, et al. Genetic ancestry analysis reveals misclassification of commonly used cancer cell lines. Cancer Epidemiol Biomarkers Prev. 2019;28(6):1003-1009, doi: 10.1158/1055-9965.EPI-18-1132.

Addressing the Disparities in Prostate Cancer Care and Outcomes

Abstract: Angiogenesis is known to play an important role in normal ovarian physiology as well as in growth and progression of ovarian cancer. The first FDA approval of bevacizumab in 2004 was for metastatic colorectal cancer in combination with chemotherapy; this was a key point for several subsequent approvals of antiangiogenic drugs. The efficacy of bevacizumab treatment is modest, however, and most ovarian cancer patients eventually develop acquired resistance, which highlights the need for new targeted therapies and/or combination strategies. Understanding the multitude of variables in response to antiangiogenic therapy would offer potential strategies for selecting patients most likely to benefit from such therapy.

The principles of antiangiogenic therapy are rooted in observations made by Folkman and others over 40 years ago.[1] Since then, antiangiogenic therapies have been incorporated into clinical care for many cancer patients. The vascular endothelial growth factor (VEGF)/VEGF receptor (VEGFR) pathway is arguably the most advanced for clinical development. This pathway has been extensively studied for its role in development of malignant ascites and tumor progression.[2] The VEGF molecule is a key angiogenic factor, which is involved in development of physiologic and pathologic neovascularization. VEGF is a growth as well as proliferation factor for endothelial cells and it binds to VEGFR-1 and VEGFR-2 as well as the cofactors neuropilin-1 (NRP-1) and NRP-2.[3] VEGFR-2 is expressed mostly on endothelial cells; VEGFR-1 can be expressed on macrophages, tumor cells, and fibroblasts.[4] While one of the main antiangiogenic approaches is to block VEGF using a monoclonal antibody, other drugs including VEGF pathway inhibitors such as small-molecule tyrosine kinase inhibitors (eg, sunitinib, sorafenib, pazopanib, regorafenib, lenvatinib, vandetanib), soluble VEGF decoy receptor (eg, aflibercept), and human monoclonal antibody against VEGFR-2 (ramucirumab) have also been developed.

Among the various antiangiogenic drugs, bevacizumab, a humanized anti-VEGF monoclonal antibody, is the most widely studied across tumor types and specifically in epithelial ovarian cancer. Beginning in 2004, the US Food and Drug Administration (FDA) granted several approvals for antiangiogenic drugs with a main indication in metastatic disease for colorectal cancer, non–small-cell lung, renal cell, cervical, and ovarian cancers, and glioblastoma. Recently, in June 2018, bevacizumab was approved for the first time also for frontline therapy for ovarian cancer in combination with platinum and taxane-based chemotherapy. Despite positive initial response rates using bevacizumab in advanced ovarian cancer, patients unfortunately develop relapsed disease within weeks to months. Nevertheless, preclinical data suggest that prolonged administration of bevacizumab as maintenance therapy after platinum-based chemotherapy prolongs progression-free survival (PFS).[5] This review summarizes the major clinical trials that led to the approval of antiangiogenic drugs and gives a brief view into novel combinations of bevacizumab with other targeted therapies in an attempt to enhance the efficacy of bevacizumab.

Ovarian cancer is the most deadly gynecologic malignancy with 5-year survival rates ranging from 47% for all stages compared to only 29% for those with advanced-stage disease (III or IV) and is responsible for approximately 14,000 deaths in the United States annually.[6] The standard-of-care treatment remains primary tumor reductive surgery followed by adjuvant platinum and taxane-based chemotherapy. However, the recurrence rate is approximately 80%, even for patients who respond to initial treatment.[7]

GOG-218 and ICON7 were the first two positive advanced-stage frontline ovarian cancer randomized phase III trials that added bevacizumab to chemotherapy.[8,9] In both GOG-218 and ICON7, bevacizumab was added to the standard of care, which consists of 6 cycles of platinum and taxane-based chemotherapy. In GOG-218, the duration of bevacizumab treatment was up to 22 cycles using a dose of 15 mg/kg, and in ICON7, 18 cycles of bevacizumab were given using a dose of 7.5 mg/kg. Although the two trials used different treatment duration and dosing, they both showed an increase in PFS. In GOG-218, the median PFS after adding bevacizumab in frontline and maintenance was increased by 4 months, from 10.3 to 14.1 months (hazard ratio (HR), 0.72; 95% CI, 0.63–0.82).[8] In ICON7, PFS at 36 months was 20.3 months with standard therapy, as compared with 21.8 months with standard therapy plus bevacizumab (HR, 0.81; 95% CI, 0.70–0.94; 

 = .004).[9] The same study suggested that the benefit of adding bevacizumab was the greatest for patients with high risk of progression (International Federation of Gynecology and Obstetrics [FIGO]) stage III > 1 cm or IV or suboptimally debulked surgery) where the estimated median PFS was 10.5 months with standard therapy, as compared with 15.9 months with bevacizumab (HR, 0.68; 95% CI, 0.55–0.85; 

 < .001).[9] Another intriguing observation of the study was that the effect on PFS was lost when the therapy with bevacizumab was withdrawn.[9,10] In this context, the optimal strategy for bevacizumab might be to continue the treatment until disease progression, as indicated by the data in the OCEANS trial in recurrent platinum-sensitive setting, where the experimental arm consisted of standard chemotherapy, gemcitabine plus carboplatin, in combination with bevacizumab as maintenance.[11] The same observation was made in the GOG-213 trial where the enrolled patients presenting recurrent platinum-sensitive disease were randomly assigned to standard paclitaxel and carboplatin every 3 weeks with or without bevacizumab. The latter was continued as maintenance every 3 weeks until disease progression or unacceptable toxicity.[12] As already mentioned from the OCEANS trial results, the effects of adding bevacizumab to chemotherapy and continuation as maintenance were reproducible. This was also the case in patients enrolled in the GOG-213 trial, where the median overall survival (OS) in the chemotherapy plus bevacizumab group was 42.2 months (95% CI, 37.7–46.2) vs 37.3 months (95% CI, 32.6–39.7) in the chemotherapy group (HR, 0.829; 95% CI, 0.683–1.005; 

 = .056).[12] Results from the ENGOT-ov15/AGO-OVAR 17 trial will most probably give an answer to the optimal treatment duration with bevacizumab, as this trial consists of 2 experimental arms, one being frontline chemotherapy combined with bevacizumab for 15 months while the second arm incorporates bevacizumab with the same frontline chemotherapy, but for a longer period (30 months). Considering the increase in PFS and OS after administering bevacizumab in frontline therapy in GOG-218 and ICON7, the European Medicines Agency (EMA) granted first approval of the drug for ovarian cancer in 2011 while the FDA approved it in June 2018.

Other Antiangiogenic Drugs Used as Upfront Therapy

Other multitarget compounds have demonstrated promising results in phase III trials. In the AGO-OVAR 16 trial, pazopanib (a protein kinase inhibitor of VEGFR and platelet-derived growth factor receptor [PDGFR]) maintenance therapy showed that PFS was higher in the experimental arm with a median improvement of 5.6 months (HR, 0.77; 95% CI, 0.64–0.91). Grade 3 or 4 hypertension (30.8%), neutropenia (9.9%), liver-related toxicity (9.4%), and diarrhea (8.2%) were significantly higher in the pazopanib arm.[13] A substudy of the same group of patients looked at the germline 

 mutation status and explored the effectiveness of maintenance therapy with pazopanib in these patients. Consistent with previous observations, germline 

 status was prognostic, demonstrating that the median PFS was longer in the placebo arm for the 

 wild-type patients (30.3 vs 14.1 months) (HR, 0.48; 95% CI, 0.29–0.78).[14] In another study, DNA from tumor and blood samples from patients enrolled in GOG-218 were sequenced, showing that patients with mutations in homologous recombination repair (HRR) had significantly prolonged PFS and OS when compared to those without mutations.[15] These data show that genetic testing plays an important role with regard to prognostic implications and should be discussed with all patients. This is particularly relevant for treatment with poly (ADP-ribose) polymerase (PARP) inhibitors.[16] There was no positive or negative correlation between HRR and the use of bevacizumab as maintenance therapy in this study.[15]

In the AGO-OVAR 12 study, another multitargeted compound, nintedanib (a potent inhibitor of VEGFR, fibroblast growth factor receptor [FGFR], and PDGFR), was investigated. Patients with advanced ovarian cancer were given standard-of-care chemotherapy, carboplatin and paclitaxel, and randomized to nintedanib or placebo. The median PFS was significantly improved in the experimental arm compared with placebo (17.2 vs 16.6 months) (HR, 0.84; 95% CI, 0.72–0.98). Patients treated with nintedanib reported grade 3 adverse effects such as diarrhea (21% vs 2%), thrombocytopenia (12% vs 5%), and anemia (12% vs 6%).[17] Another trial with trebananib, a selective angiopoietin-1 (Ang1) and Ang2 neutralizing peptibody, examined its effects in combination with standard chemotherapy as well as maintenance. This trial was a randomized study, in which patients received trebananib or placebo plus carboplatin and paclitaxel as first-line treatment. The primary endpoint of the study, PFS, did not show any improvement after the combination of trebananib with carboplatin and paclitaxel.[18]

Bevacizumab and Other Antiangiogenic Drugs Used in the Relapse Setting, Platinum Resistant

While the antiangiogenic frontline therapy was seen from different perspectives by the FDA and the EMA, bevacizumab was initially granted FDA approval for use in recurrent ovarian cancer based on the AURELIA trial. The experimental arm of this trial consisted of adding bevacizumab to chemotherapy until disease progression. Patients who received bevacizumab had a median PFS of 6.7 vs 3.4 months in patients receiving chemotherapy alone (HR, 0.48; 95% CI, 0.38–0.60) and median OS of 16.6 vs 13.3 months, respectively (HR, 0.85; 95% CI, 0.66–1.08).[19] Another trial, TRINOVA-1, aimed to determine whether adding trebananib to weekly paclitaxel until disease progression would improve PFS. This trial showed that inhibition of Ang1/Ang2 determined a prolongation of PFS but not over the benefit of bevacizumab.[20] A third randomized phase III trial, ENGOT-ov-6/TRINOVA-2, which had a similar study design to TRINOVA-1, evaluated PFS and OS after adding trebananib to pegylated liposomal doxorubicin (PLD) in patients with recurrent ovarian cancer (platinum sensitive and resistant). Patients in the experimental arm had a median PFS of 7.6 months (95% CI, 7.2–9.0) vs 7.2 months (95% CI, 4.8–8.2) in the placebo arm (HR, 0.92; 95% CI, 0.68–1.24). Trebananib demonstrated anticancer activity indicated by improved objective response rate (46% vs 21% in the placebo arm [odds ratio, 3.43; 95% CI, 1.78–6.64]). Median duration of response was also improved in the trebananib arm (7.4 months; 95% CI, 5.7–7.6) vs placebo (3.9 months; 95% CI, 2.3–6.5).[21]

Bevacizumab and Other Antiangiogenic Drugs Used in the Relapse Setting, Platinum Sensitive

In patients with recurrent platinum-sensitive ovarian cancer, the OCEANS and GOG-213 trials led to important advances. In the OCEANS trial, bevacizumab or placebo was added to gemcitabine and carboplatin for 6 to 10 cycles and then continued until progression. The addition of bevacizumab led to a statistically significant increase in PFS (HR, 0.484; 95% CI, 0.388–0.605). Median PFS was 12.4 months for bevacizumab vs 8.4 months in the placebo group.[11] In the GOG-213 trial, a different chemotherapy backbone (paclitaxel and carboplatin) was used and it was not placebo-controlled due to its primary endpoint being OS. In addition, surgical randomization for candidates deemed appropriate for secondary surgery was done for about 16% of the intent-to-treat population. This trial also demonstrated an increase in the median PFS in the bevacizumab plus chemotherapy arm compared with chemotherapy alone (13.8 months vs 10.4 months). Of note, GOG-213 showed a positive trend in median OS from 37.3 months in the chemotherapy group to 42.2 months in the chemotherapy plus bevacizumab group (HR, 0.829; 95% CI, 0.683–1.005; 

 = .056), which when adjusted for platinum-free interval and participation in the surgical objective, was significantly associated with OS (HR, 0.823; 95% CI, 0.680–0.996, 

 = .0447).[12] While the patients in the OCEANS trial did not receive bevacizumab with the frontline treatment, a third trial, MITO16B-MaNGO OV2B-ENGOT OV17, enrolled patients with carboplatin-sensitive relapsed disease who had had previous bevacizumab treatment. They found that bevacizumab in combination with a platinum-based doublet is associated with significantly prolonged PFS and a good safety profile.[22] Results from AGO OVAR 2.21, which evaluated patients with relapsed disease who previously had bevacizumab in frontline, were presented during the ESMO 2018 meeting. In this study, both arms contained bevacizumab added to a platinum-based chemotherapy; one arm consisted of gemcitabine and carboplatin and the other arm consisted of PLD and carboplatin. Being the first phase III trial in ovarian cancer comparing two bevacizumab-containing regimens, the study showed that the carboplatin/PLD/bevacizumab combination resulted in superior PFS compared with carboplatin/gemcitabine/bevacizumab (HR, 0.80; 95% CI, 0.68–0.96; 

Only one antiangiogenic VEGF receptor 1-3 inhibitor investigated in recurrent platinum-sensitive ovarian cancer patients so far has shown anti-tumor activity. In the 3-arm ICON6 trial, participants were randomly assigned to receive placebo plus chemotherapy then placebo maintenance (arm A), cediranib 20 mg once daily plus chemotherapy then placebo maintenance (arm B), or cediranib 20 mg once daily plus chemotherapy then cediranib 20 mg once daily for maintenance (arm C). Median PFS was 11 months for arm C vs 8.7 months for arm A (HR, 0.56; 95% CI, 0.44–0.72). The difference in OS was nonsignificant (

 = .3) across the 3 arms.[24] Although the improvement in PFS to a median of 11 months with cediranib as maintenance is similar to that reported with bevacizumab maintenance after chemotherapy, the risk of hypertension, fatigue, diarrhea, and nausea in patients taking cediranib in combination with chemotherapy makes this a worse candidate compared to intravenous bevacizumab.

Other Emerging Antiangiogenic Therapies in Combination With Anti-VEGF Drugs

Using antiangiogenic drugs to target the VEGF molecule or angiogenic receptors such as VEGFR, FGFR, and PDGFR proved to be an effective strategy for inhibiting tumor angiogenesis. The effects on PFS and OS are modest, as most patients develop adaptive resistance to anti-VEGF therapy and eventually present with progressive disease. Novel concepts that combine antiangiogenic drugs with other emerging approaches such as Delta-like ligand 4 (DLL4)-blocking antibody,[25,26] and colony stimulating factor 1 receptor (CSF1R) found on macrophages,[27] are showing promising results in decreasing tumor burden in vivo.[25-27] Previous studies that investigated the DLL4 pathway showed that DLL4 blockade can inhibit tumor growth by promoting nonproductive angiogenesis.[25] Furthermore, dual targeting of VEGF and DLL4 showed superior anti-tumor effects (decreased tumor burden, less ascites) in ovarian cancer mouse models compared with either therapy alone.[26] Navicizixumab, a DLL4 and VEGF inhibitor, is a clinical candidate being tested in a phase I/Ib study.[28]

With the genetic ablation of CSF1R and thereby macrophages, it was observed that the macrophages could have a protumoral effect.[29,30] Additional studies have shown that macrophages support tumor growth by production of proangiogenic factors as VEGF and other angiogenic chemokines (eg, CXCL12).[31] After emergence of resistance to anti-VEGF therapy and paclitaxel in an in vivo mouse model, CSF1R inhibitor therapy added to anti-VEGF therapy restored response to antiangiogenic therapy and an 83% lower tumor burden compared to treatment with anti-VEGF therapy and paclitaxel alone.[27]

Phase III clinical trials showed that adding PARP inhibitors in maintenance therapy improves PFS for patients with recurrent epithelial ovarian, fallopian-tube, or primary peritoneal cancer who had received prior platinum-based chemotherapy.[32,33] The SOLO-1 study demonstrated that the risk of disease progression or death was 70% lower with olaparib than with placebo (HR, 0.3; 95% CI, 0.23–0.41; 

 < .001) in patients with newly diagnosed advanced ovarian cancer who have a complete or partial response after platinum-based chemotherapy.[16] Taking into account the response rate to PARP inhibitors, the combination of PARP inhibitor with bevacizumab is the premise of the PAOLA-1 phase III clinical trial. Specifically, olaparib and bevacizumab are used in frontline maintenance in advanced ovarian cancer. Different combinations of PARP inhibitors and antiangiogenic drugs are studied in the NRG-GY005 (NCT02502266) phase II/III and NRG-GY004 (NCT02446600) phase III trials in which a PARP inhibitor is given alone or in combination with cediranib maleate in the experimental arms to determine efficacy compared with standard platinum-based chemotherapy in platinum-sensitive relapsed ovarian cancer.

VEGF inhibitors in combination with immunotherapy are also being investigated. The rationale for combining immunotherapy with VEGF inhibitors is that the latter induce abundant tumor vasculature and inhibit directly T-cell function.[34] There are several promising trials in development evaluating immunotherapy in combination with VEGF inhibitors such as NRG-GY009 (NCT02839707), ATLANTE (NCT02891824), and IMagyn050 (NCT03038100).

Possible Markers of Response to Antiangiogenic Therapy

Treatment with antiangiogenic agents can be quite expensive and can have substantial side effects. One potential solution for optimizing the treatment is to identify useful biomarkers that can allow selection of patients most likely to benefit from such therapy and potentially to detect early escape from anti-VEGF therapy. Circulating molecular and cellular biomarkers in blood would be ideal candidates, as such markers can be followed longitudinally. However, circulating protein levels (eg, VEGF or other angiogenic molecules such as placental growth factor [PIGF]) have not been particularly effective.[35]

Circulating endothelial cells (CECs) are mobilized in murine models as well as in humans in response to VEGF.[36,37] This is especially important because the level of CECs should decrease in response to anti-VEGF therapy. Although the suppression of mobilized endothelial cells was observed in patients with confirmed response to anti-VEGF therapy, an increase in value before clinical progression was not noted.[38]

Some have attempted to identify molecular subgroups of high-grade serous ovarian cancer that could predict response to anti-VEGF therapy. For example, Gourley and colleagues identified three major subgroups in the ICON7 participants based on mRNA analysis of human tissue samples. In two of the subgroups, the angiogenic genes were upregulated, while the third subgroup presented more complex features including repression of the angiogenic genes and upregulation of the immune genes. With regard to OS, the latter group had a superior OS compared to the other two subgroups. However, addition of bevacizumab therapy in this latter group was associated with worse PFS and OS compared to chemotherapy alone. This study suggests that an immune subgroup “geno-signature” with the repressed angiogenic-related expression could better select patients in whom antiangiogenic therapy could be avoided. Addition of bevacizumab in the proangiogenic group was associated with a nonsignificant trend for improved PFS compared with controls (median, 17.4 vs 12.3 months, respectively).[39]

Given the lack of reliable plasma biomarkers, imaging modalities such as CT and MRI scans are the most common assessment tools to evaluate efficacy. However, other functional imaging methods such as dynamic contrast-enhanced (DCE)-MRI and CT perfusion can provide information on tumor perfusion, vascularity, and permeability and could be useful as early markers of treatment efficacy.[38,40,41] The ACRIN 6695 study investigated whether CT perfusion biomarkers are associated with PFS in patients with advanced ovarian cancer treated with carboplatin and dose-dense or conventional paclitaxel. The patients were enrolled from the prospective phase III GOG-0262 clinical trial in which bevacizumab use was left to the physician’s choice and was overwhelmingly preferred based on the results from previous clinical trials in patients with advanced ovarian cancer. The study showed that changes in vasculature parameters within 4 weeks of initiating therapy in one or more CT-perfusion biomarkers were associated with shorter PFS, proving that these biomarkers can be considered good imaging candidates to provide early prognostic information.[41]

The number of antiangiogenic drugs developed during the last 30 years is impressive and is associated with improved patient outcomes. Nevertheless, due to emergence of adaptive resistance, additional therapies are needed. Steps forward would be made with the identification of predictive biomarkers that could help assess the response among patients and a better understanding of mechanisms of adaptive resistance. While neutralization of VEGF was the first approach for targeting the tumor microenvironment, additional approaches (eg, immune therapy) have been also quite successful. In this context, the potential of tumor vessel normalization using antiangiogenic drugs to improve the effectiveness of immunotherapy should be explored further. Collectively, such microenvironment-targeted approaches hold promise for further improving the outcomes of women with ovarian cancer.

Dr. Chelariu-Raicu has no significant financial interest in or other relationship with the manufacturer of any product or provider of any service mentioned in this article. Dr. Sood has received funding from Merck, Kiyatec, and Bio-Path. Dr. Coleman has been a consultant, speaker, or advisor for AbbVie, Cell Medica, Genmab, MORE Health, Takeda, Roche, Merck, Aravive, ArQule, AstraZeneca, Clovis, Eisai, Genentech, Immunogen, and Janseen

ncorporating Antiangiogenic Agents in the Treatment of Advanced Ovarian Cancer

The incorporation of antiangiogenic therapy in the treatment of solid tumor malignancies dates back to 2004, when bevacizumab, an anti-vascular endothelial growth factor (VEGF) monoclonal antibody, received US Food and Drug Administration (FDA) approval in advanced colorectal cancer after demonstrating an improvement in both overall and progression-free survival (PFS).[1] Studies in epithelial ovarian cancer have shown a correlation between VEGF expression and more advanced disease leading to worse survival rates.[2] After initial skepticism, bevacizumab was finally granted FDA approval in 2011 for recurrent ovarian cancer and has since received approval in 2018 as front-line therapy after initial surgical resection.[3] Other antiangiogenic agents of interest include pazopanib, nintedanib, cediranib, and navicizumab, all of which are currently being evaluated in clinical trials.

Although bevacizumab demonstrated a modest improvement in PFS, the unique toxicity profile of antiangiogenic therapies and the lack of an overall survival benefit must be considered when selecting patients most appropriate for treatment. Common adverse effects can be divided into these major categories: generalized, cardiovascular, gastrointestinal (GI), and dermatologic. Generalized symptoms primarily include fatigue, headache, dyspnea, arthralgia, and myalgia.[4] In the phase III PALETTE trial that evaluated using pazopanib in treating advanced soft-tissue sarcoma, the incidence of all-grade and grade > 3 fatigue was 65% and 13%, respectively.[5] Although not FDA approved for use in ovarian cancer, the debilitating fatigue associated with pazopanib often warrants dose modification to improve tolerability. Cardiovascular symptoms such as hypertension and venous thromboembolism can be concerning in a primarily postmenopausal population. Frequent blood pressure (BP) monitoring is essential when treating patients with baseline hypertension, and all patients should have adequate BP control prior to initiating treatment with anti-VEGF therapies. Common GI symptoms include diarrhea and abdominal pain. More severe GI adverse events (ie, perforation, fistula, and/or necrosis) have been reported in 2.9% of patients with ovarian cancer, and identifiable risk factors include history of inflammatory bowel disorder and previous bowel resection.[6] Palmar-plantar erythrodysesthesia has been reported in patients receiving anti-VEGF therapies, and early identification of symptoms is essential in preventing dose interruptions. Impaired wound healing can result in serious or fatal complications; therefore, anti-VEGF therapies should be held at least 28 days prior to and after elective invasive procedures.[4] Considering the potentially substantial toxicities associated with bevacizumab and other anti-VEGF agents, their incorporation into the standard treatment of ovarian cancer should be reserved for patients with high-risk clinical features and minimal comorbidities.

Dr. Byers has served on advisory boards for Pfizer, Tesaro, and Heron Therapeutics, and participated in surveys for The Delham Group

 Genentech press release. FDA approves Avastin, a targeted therapy for first-line metastatic colorectal cancer patients. February 26, 2004. Available at: 

https://www.gene.com/media/press-releases/7167/2004-02-26/fda-approves-avastin-a-targeted-therapy

 Duncan TJ, Al-Attar A, Rolland P, et al. Vascular endothelial growth factor expression in ovarian cancer: a model for targeted use of novel therapies. Clin Cancer Res. 2008;14:3030-5.

 US Food and Drug Administration. FDA approves bevacizumab in combination with chemotherapy for ovarian cancer. June 13, 2018. Available at: 

www.fda.gov/Drugs/InformationOnDrugs/ApprovedDrugs/ucm610664.htm

 Avastin (bevacizumab) [prescribing information]. South San Francisco, CA: Genentech, Inc.; February 2019.

 van der Graaf WT, Blay JY, Chawla SP, et al; EORTC Soft Tissue and Bone Sarcoma Group; PALETTE study group. Pazopanib for metastatic soft-tissue sarcoma (PALETTE): a randomized, double-blind, placebo-controlled phase 3 trial. Lancet. 2012;379:1879-86.

 Burger RA, Brady MF, Bookman MA, et al. Risk factors for GI adverse events in a phase III randomized trial of bevacizumab in first-line therapy of advanced ovarian cancer: a gynecologic oncology group study. J Clin Oncol. 2014;32:1210-7.

 Folkman J. Tumor angiogenesis: Therapeutic implications. N Engl J Med. 1971; 285:1182-6.

 Graybill W, Sood AK, Monk BJ, Coleman RL. State of the science: emerging therapeutic strategies for targeting angiogenesis in ovarian cancer. Gynecol Oncol. 2015;138:223-6.

 Carmeliet P, Jain RK. Molecular mechanisms and clinical applications of angiogenesis. Nature. 2011;473:298-307.

 Fischer C, Mazzone M, Jonckx B, Carmeliet P.FLT1 and its ligands VEGFB and PlGF: drug targets for anti-angiogenic therapy? Nat Rev Cancer. 2008;8:942-56.

Mabuchi S, Terai Y, Morishige K, et al. Maintenance treatment with bevacizumab prolongs survival in an in vivo ovarian cancer model. Clin Cancer Res. 2008;14:7781-9.

Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin. 2019;69:7-34.

Monk BJ, Herzog TJ, Tewari KS. Evolution of chemosensitivity and resistance assays as predictors of clinical outcomes in epithelial ovarian cancer patients. Curr Pharm Des. 2016;22:4717-28.

Burger RA, Brady MF, Bookman MA, et al. Gynecologic Oncology Group. Incorporation of bevacizumab in the primary treatment of ovarian cancer. N Engl J Med. 2011;365:2473-83.

Perren TJ, Swart AM, Pfisterer J, et al. A phase 3 trial of bevacizumab in ovarian cancer. N Engl J Med. 2011;365:2484-96.

Oza AM, Cook AD, Pfisterer J, et al. Standard chemotherapy with or without bevacizumab for women with newly diagnosed ovarian cancer (ICON7): overall survival results of a phase 3 randomised trial. Lancet Oncol. 2015;16:928-36.

Aghajanian C, Blank SV, Goff BA, et al. OCEANS: a randomized, double-blind, placebo-controlled phase III trial of chemotherapy with or without bevacizumab in patients with platinum-sensitive recurrent epithelial ovarian, primary peritoneal, or fallopian tube cancer. J Clin Oncol. 2012;30:2039-45.

Coleman RL, Brady MF, Herzog TJ, et al. Bevacizumab and paclitaxel-carboplatin chemotherapy and secondary cytoreduction in recurrent, platinum-sensitive ovarian cancer (NRG Oncology/Gynecologic Oncology Group study GOG-0213): a multicentre, open-label, randomised, phase 3 trial. Lancet Oncol. 2017;18:779-91.

du Bois A, Floquet A, Kim JW, et al. Incorporation of pazopanib in maintenance therapy of ovarian cancer. J Clin Oncol. 2014;32:3374-82.

Harter P, Johnson T , Berton-Rigaud D, et al. BRCA1/2 mutations associated with progression-free survival in ovarian cancer patients in the AGO-OVAR 16 study. Gynecol Oncol. 2016;140:443-9.

Norquist BM, Brady MF, Harrell MI, et al. Mutations in homologous recombination genes and outcomes in ovarian carcinoma patients in GOG 218: an NRG Oncology/Gynecologic Oncology Group Study. Clin Cancer Res. 2018;24:777-83.

Moore K, Colombo N, Scambia G, et al. Maintenance olaparib in patients with newly diagnosed advanced ovarian cancer. N Engl J Med. 2018;379:2495-505.

du Bois A, Kristensen G, Ray-Coquard I, et al. Standard first-line chemotherapy with or without nintedanib for advanced ovarian cancer (AGO-OVAR 12): a randomised, double-blind, placebo-controlled phase 3 trial. Lancet Oncol. 2016;17:78-89.

Vergote I, Scambia G, O’Malley DM, et al. Trebananib or placebo plus carboplatin and paclitaxel as first-line treatment for advanced ovarian cancer (TRINOVA-3/ENGOT-ov2/GOG-3001): a randomised, double-blind, phase 3 trial. Lancet Oncol. 2019;20:862-76.

Pujade-Lauraine E, Hilpert F, Weber B, et al. AURELIA: A randomized phase III trial evaluating bevacizumab (BEV) plus chemotherapy (CT) for platinum (PT)-resistant recurrent ovarian cancer (OC). J Clin Oncol. 2012;30(suppl):abstr LBA5002.

Monk BJ, Poveda A, Vergote I, et al. Anti-angiopoietin therapy with trebananib for recurrent ovarian cancer (TRINOVA-1): a randomised, multicentre, double-blind, placebo-controlled phase 3 trial. Lancet Oncol. 2014;15:799-808.

Marth C, Vergote I, Scambia G, et al. ENGOT-ov-6/TRINOVA-2: Randomised, double-blind, phase 3 study of pegylated liposomal doxorubicin plus trebananib or placebo in women with recurrent partially platinum-sensitive or resistant ovarian cancer. Eur J Cancer. 2017;70:111-21.

Pignata S, Lorusso D, Joly F. et al Chemotherapy plus or minus bevacizumab for platinum-sensitive ovarian cancer patients recurring after a bevacizumab containing first line treatment: the randomized phase 3 trial MITO16B-MaNGO OV2B-ENGOT OV17. J Clin Oncol. 2018;36(suppl):abstr 5506.

Pfisterer J, Dean AP, Baumann K. et al. Carboplatin/pegylated liposomal doxorubicin/bevacizumab (CD-BEV) vs. carboplatin/gemcitabine/bevacizumab (CG-BEV) in patients with recurrent ovarian cancer. A prospective randomized phase III ENGOT/GCIG-Intergroup study (AGO Study Group, AGO-Austria, ANZGOG, GINECO, SGCTG). Ann of Oncol. 2018:29(suppl 8):abstr 9330.

Ledermann JA, Embleton AC, Raja F, et al. Cediranib in patients with relapsed platinum-sensitive ovarian cancer (ICON6): a randomised, double-blind, placebo-controlled phase 3 trial. Lancet. 2016; 387:1066-74.

Noguera-Troise I, Daly C, Papadopoulos NJ, et al. Blockade of Dll4 inhibits tumour growth by promoting non-productive angiogenesis. Nature. 2006;444:1032-1037.

Huang J, Hu W, Hu L, et al. Dll4 inhibition plus aflibercept markedly reduces ovarian tumor growth. Mol Cancer Ther. 2016;15:1344-52.

Lyons YA, Pradeep S, Wu SY, et al. Macrophage depletion through colony stimulating factor 1 receptor pathway blockade overcomes adaptive resistance to anti-VEGF therapy. Oncotarget. 2017;8:96496-505.

Jimeno A, Morre KN, Gordon M, et al. A first-in-human phase 1a study of the bispecific anti-DLL4/anti-VEGF antibody navicixizumab (OMP-305B83) in patients with previously treated solid tumors. Invest New Drugs. 2019;37:461-72.

Lin EY, Nguyen AV, Russell RG, Pollard JW. Colony-stimulating factor 1 promotes progression of mammary tumors to malignancy. J Exp. Med. 2001;193,727-40.

Lin EY, Pollard JW. Tumor-associated macrophages press the angiogenic switch in breast cancer. Cancer Res. 2007;67:5064-6.

Hughes R, Qian BZ, Rowan C, et al. Perivascular M2 macrophages stimulate tumor relapse after chemotherapy. Cancer Res. 2015;75:3479-91.

Mirza MR, Monk BJ, Herrstedt J, et al. Niraparib maintenance therapy in platinum-sensitive, recurrent ovarian cancer. N Engl J Med. 2016;375:2154-64.

Coleman RL, Oza AM, Lorusso D, et al. Rucaparib maintenance treatment for recurrent ovarian carcinoma after response to platinum therapy (ARIEL3): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet. 2017;390:1949-61.

Yang J, Yan J, Liu B. Targeting VEGF/VEGFR to modulate antitumor immunity. Front Immunol. 2018;9:978.

Jain RK, Duda DG, Willett CG, et al. Biomarkers of response and resistance to antiangiogenic therapy. Nat Rev Clin Oncol. 2009;6:327-38.

This review summarizes the major clinical trials that led to the approval of antiangiogenic drugs for ovarian cancer and gives a brief view into novel combinations of bevacizumab with other targeted therapies in an attempt to enhance the efficacy of bevacizumab.

Anti-Angiogenesis Therapy in Ovarian Cancer: Which Patient is It Most Likely to Benefit? 

Throughout the last half of the 20th century, despite enormous research efforts, lung cancer remained the most feared as well as lethal cancer across the world. Starting in 1991, a multidisciplinary group of investigators (ELCAP) at Weill Cornell Medical College (WCMC) in New York, mindful of challenge, proposed to leverage the transformative breakthrough in performance of low-dose radiation CT imaging to obtain full thoracic images in a single breath hold to identify early, curable lung cancer. Their research demonstrated that it worked remarkably well.[1] However, they were to find convincing a skeptical world already inured by decades of failed lung cancer detection efforts was to be their greatest challenge.

The research team wanted to learn about previous efforts. For this purpose, they conducted extensive discussions with veterans of the previous, ill-fated National Cancer Institute (NCI)-sponsored randomized trial evaluating the effectiveness of sputum cytology and annual chest x-rays (CXR) as an early detection method.[2] A particularly insightful expert was the chief study statistician, B. Flehinger. Dr. Flehinger had developed a mathematical model to estimate the potential cure rates for lung cancers that were much smaller than were detectable by CXR, but in a range that could be detected by low-dose CT (LDCT).[3,4] Based on the preliminary experience, the investigators hypothesized that the LDCT detection rate of small lung cancer could be as high as 80% of cancers between 4 mm and 10 mm and the Flehinger model suggested that for such cancers, the cure rate could be as high as 80%, much higher than the 18% rate the model predicted for CXR screening.

Based on this preliminary WCMC modeling experience, early detection of curable lung cancer using LDCT seemed feasible. The investigators understood that to compare the quantitative benefit of CT to CXR screening, three probabilities need to be determined for each imaging modality:

1) Probability of detecting a noncalcified nodule (NCN) in a person

2) Probability of diagnosing lung cancer by size and stage

3) Probability of curing all persons diagnosed with lung cancer under screening

They also recognized that, for each imaging modality, these three probabilities need to be determined separately for the baseline round and for all subsequent repeat rounds of screening.[1,5,6] The profound difference between the probabilities for the baseline (“prevalence”) round and the subsequent rounds of repeat screening (“incidence”) had been recognized by Morrison and others.[7] The terms “prevalence” and “incident” had been adapted from acute disease descriptors (eg, myocardial infarction), but any cancer has typically been present but not detectable for a considerable time and thus is a prevalent case, whether detected on baseline or repeat screening. Thus, the ELCAP investigators do not use the terms prevalence or incident cancers, but rather report whether the cancer was detected in the baseline or repeat rounds.

While the probabilities for the baseline round are different from those of repeat rounds of screening, each repeat annual round is a replicate of the other repeat rounds. Thus, a single repeat round would suffice to provide the probabilities, provided the sample size was sufficiently large. The investigators decided that the screening interval should be annual,[5,6] based on analysis of the lung cancer growth rates of lung cancers.[8]

The group’s first study was called the Early Lung Cancer Action Program (ELCAP).[1,6] Its goal was to determine the first two probabilities for LDCT and CXR by recruiting a prospective cohort of 1,000 current and former cigarette smokers, aged 60 years or older with at least 10 pack-years of cigarette smoking, who were asymptomatic for lung cancer. Each ELCAP participant was to receive a baseline LDCT and CXR followed by a repeat screening 12 months later. Upon receiving Institutional Review Board approval in 1992 at WCMC and then at New York University Medical Center, ELCAP recruitment started, thanks to seed philanthropic funding. A grant application was submitted to the NCI, but it was recognized that this would probably take several years to receive a grant award, and as predicted it was awarded in 1995.[5]

During the conduct of the ELCAP trial, participants with noncalcified nodules (NCNs) on LDCTs were presented at weekly multidisciplinary Thoracic Tumor Conferences. Quarterly ELCAP meetings of investigators at both institutions were held to rigorously manage the conduct of this trial. The literature at that time focused on “solitary noncalcified pulmonary nodules,”[9,10] but it was soon recognized that on LDCT, multiple small NCNs were frequently detected in a single participant, and more importantly only few of them represented lung cancers. Thus the focus changed from detecting solitary NCN to detecting one or more NCNs. More importantly, it was recognized that most NCNs remained stable while the lung cancer was identified in rapidly growing LDCT-detected NCNs.[11] Furthermore, standard diagnostic testing and follow-up of NCNs was based on CXR experience, but the ELCAP investigators rapidly moved away from performing a biopsy on all NCNs. Instead for NCN less than 10 mm in size on baseline LDCT, follow-up CT in 3 months was performed to assess growth.[1] If, on follow-up LDCT, the observed NCN growth rate was suggestive of a malignancy, an immediate nodule biopsy was recommended. For larger NCNs (≥ 10 mm) the standard recommendation, which called for biopsy, was followed.

Dr. D. Yankelevitz, an ELCAP investigator, recognized the potential of CT to reliably measure growth rates of small lung cancers on follow-up CTs and led the research projects to confirm that short-term follow-up LDCT scans were very useful in distinguishing between the many small benign NCNs and the few malignant ones.[11-16] This critical innovation was initially supported by philanthropy but then subsequently validated by NCI grant awards,[17] a textbook example of the power of public/private interactions.

The seminal finding of using short-term follow-up LDCT for differentiating lung cancer from benign nodules also highlighted how the efficiency of screening diagnostic workup could be optimized and how it could be integrated into protocols.[18,19] This innovation was rapidly recognized and also integrated by a wide range of other screening research groups.[20,21] In particular, the NELSON trial in the Netherlands and Belgium integrated volume assessment into their protocol.[22]

With the large number of participants with NCNs on LDCT with the need for follow-up scheduling of CT scans, the investigators recognized that data and image management soon became overwhelming, even for a study of 1,000 participants at two institutions. Fortunately, the principal investigator, C. Henschke, was an experienced computer programmer and statistician, and was thus able to rapidly develop and write the needed computer management programs to document, store, and analyze the data for the ELCAP study.

In 1998, recruitment of 1,000 baseline participants was completed and the baseline results were analyzed and published in the 

 in 1999.[1] Among the 1,000 participants, NCNs were found by LDCT in 233 (23.3%) and by CXR in 68 (6.8%). Lung cancers were diagnosed in 27 by LDCT and 7 by CXR among the 1,000 participants (2.7% vs 0.7%, 

 = .0005). Of the 27 lung cancer patients identified by LDCT, 23 (85.1%) were stage I lung cancers while CXR identified only 4, missing 19 (82%) of them. LDCT also found 4 additional cancers, 2 endobronchial cancers, and 2 mediastinal cancers that were not identified by CXR.

Using published long-term survival rates by stage.[23,24] the ELCAP investigators projected a possible cure rate of 80% (a lung cancer fatality rate of 20%) for patients with lung cancer diagnosed in a program of LDCT screening. To robustly estimate the cure rate, however, additional screenings and longer follow-up time were needed.

The annual repeat screenings were completed, analyzed, and submitted in 1999, but only published in 

 in 2001.[6] As annual rounds are repeated year after year for each participant, while the baseline round is performed only once for each participant, results from annual screening rounds are critical to assess screening performance.

Among the 1,184 annual repeat screening, new or growing NCNs were identified in 30 (2.5%) participants. Seven of the 30 participants were diagnosed with lung cancer, of which 6 (85.7%) had stage I disease. There were far fewer new NCNs on annual repeat screening than on the baseline screening (2.5% vs 23%, 

 < .0001) and the percentage of screen-diagnosed lung cancers was also different (0.59% vs 2.7%, 

 < .0001). The percentage of screen-diagnosed stage I disease, however, was similar (85.7% vs 85.1%, 

 = 1.00), although the median tumor size was smaller in annual rounds of screening (8 mm vs 10 mm, 

 = .45). Based on this evidence and pathologic findings,[25,26] future protocols for workup in the baseline and annual repeat rounds of screening were different.[18,19,27]

A critical quality assessment parameter for any cancer screening program is the frequency of cancers identified by symptom-prompted workup between the scheduled rounds of repeat screening. Another 2 participants had symptom-prompted workup for pneumonia which led to the diagnosis of lung cancer prior to the first annual round of screening. Both had endobronchial tumors, one stage I small-cell carcinoma and one stage II squamous-cell carcinoma. Thus, the percentage of patients with stage I lung cancer, including screen-diagnosed and symptom-prompted diagnoses, was slightly lower on annual repeat than on baseline screening (77.8% vs 85.1%, 

 = .63). This was expected as a higher percentage of aggressive cancers are identified in repeat rounds of screening.[26,27]

Prior to 2000, radiologists reviewed the CT images displayed on radiographic films that provided 12 CT images per film.[1,7] Typically, 30 CT images were obtained for each screening participant, using 10-mm slice thickness, although reconstruction at 5-mm overlapping intervals was used to improve the image resolution. With advances in CT technology, particularly with multislice scanners, submillimeter slice thickness acquisitions are now obtained, providing as many as 800 CT images for each participant, and these images are displayed on large computer monitors, which, though expensive, provide far greater detail. Today, in addition to viewing axial images, coronal, sagittal, and maximum intensity projections are routinely provided, further enhancing visualization. These advances have markedly increased identification of tiny NCNs, thus reducing symptom-prompted diagnoses between annual rounds of screening. These marked improvements have also prompted reconsideration of using longer intervals between repeated rounds of screening.[21,28]

Expansion of ELCAP Screening to New York [NY]-ELCAP and International [I]-ELCAP

Immediate international requests for collaborations on LDCT screening followed the publication of the 1999 

 article.[1] In response, the ELCAP investigators initiated an International Conference on Screening for Lung Cancer in the fall of 1999 and these conferences have been continued, with the 40th conference held in April 2019.[29,30] At the first conference, participants called for the development of a multi-institutional protocol that, when presented at the next conference 6 months later, was unanimously adopted and then published.[18,19]

The screening protocol was critical, as with the marked advances of CT scanner and viewing technology, radiologists could detect NCNs as small as 2 mm but very few of these NCNs are lung cancers. The exquisite sensitivity of LDCT to identify small lung cancers, compared to CXR, further accentuated the difference in the first two probabilities of the baseline and annual repeat rounds and the need for recognition of these differences in the baseline and repeat protocol.[26,27] These advances also highlighted the need to update the threshold criteria for workup on baseline and on annual repeat screening.[18,31-33] Many additional insights were developed using the enlarging International [I]-ELCAP database, and these recommendations have been incorporated in the updated online protocol.[19] These recommendations are provided for findings made on screening scans of mediastinal masses, emphysema, coronary artery calcifications, aortic valve calcifications, pulmonary artery enlargement, interstitial lung disease, breast density and masses, liver density, and bone. In developing these recommendations, the investigators worked closely with clinicians in different medical specialties and stressed that, different from routine medical practice, these findings were not prompted by symptoms but identified as a result of screening and thus recommendations and workup need to be tailored accordingly. Some of these recommendations have been incorporated, for example, into national guidelines.[34] The importance of using a protocol with detailed recommendations was highlighted by the comparison of I-ELCAP and National Lung Screening Trial (NLST) results,[35] which showed that the use of a regimen of screening markedly decreased unnecessary workup.

In view of the international demand for LDCT screening, A.P. Reeves, a Professor of Electrical Engineering and Computer Science at Cornell University, who had been working with the ELCAP team on tumor growth analyses,[12-18] suggested expansion to a web-based management system that would enable more efficient and comprehensive data and images collection from any site connected to the internet. It would also be able to provide a platform for screening management tools, training, and reporting. Thus, the original ELCAP management system was converted into the updated, web-based ELCAP management system.[36]

This web-based ELCAP management system together with the I-ELCAP protocol made it possible to rapidly expand LDCT screening worldwide.[19] ELCAP investigators developed educational sessions for radiologists and coordinators, and provided the ELCAP management system, free of charge, to all participating institutions. Core features of this system have been widely adopted by other screening data systems. The ELCAP management system has developed a comprehensive, new site onboarding process using a blend of web content, face-to-face interactions at its biannual workshop process, as well as a comprehensive archive of relevant screening publications.[19] An example of the onboarding process is the dual readings service allowing each new participating institution to have the first 200 scans read jointly by the coordinating center with follow-up adjudication discussions and then review of the next 300 scans.[37] Other quality interactions around lung cancer pathology review, assessment of other thoracic detected disease conditions, as well as nursing and navigation measures are available.

By 2000, LDCT screening rapidly expanded throughout New York State as NY-ELCAP and internationally as I-ELCAP programs.[38,39]

I-ELCAP Long-Term Survival for Diagnosed Lung Cancers Under Screening

The expansion of ELCAP provided sufficient data to robustly estimate the probability of cure for patients diagnosed with lung cancer under lung cancer screening (baseline, annual screen-diagnoses as well as symptom-prompted diagnoses rounds).[39] The cure rate, estimated by the 10-year Kaplan-Meier survival rate of 484 patients, was 88% (95% CI, 84%–91%), regardless of the cancer stage or treatment. Patients who were diagnosed with clinical stage I lung cancer and resected within 1 month of diagnosis had a higher estimated cure rate (92% [95% CI, 88%–95%]).

These estimated cure rates for annual LDCT screening for lung cancer have been confirmed at the international conferences, held every 6 months since 1999,[29] including the latest 40th conference held in April 2019, which provided the 20-year follow-up results.

The benefit of LDCT screening was individualized as to the age, smoking history, and other pertinent information of each individual considering screening.[40] Using the ELCAP baseline information together with the long-term Kaplan-Meier survival rates, the benefit of a single round of screening was determined based on the person’s age and smoking history. This benefit was due to the increased survival by undergoing the first round of LDCT screening. This approach can be extended to calculate the individual benefit or survival gain from each subsequent annual round of screening.

The ELCAP management system provided a scalable system to disseminate the standardized data elements needed for consistent, high-quality screening implementation and management. It documented the initial contact of each participant and allowed for all subsequent steps of the screening process, scheduling, reading the LDCT, recommendations for follow-up, and follow-up. This information is needed by coordinators, navigators, nurses, radiologists, and treating physicians on a participant-prompted, timely basis. The system efficiently collected screening data obtained in the process of a clinical screening program and also provided quality control, follow-up management, and reminders. Beyond the clinical needs, the system enabled collection of screening research data, collected in the context of clinical care, for rigorous scrutiny of needed updates and analysis of outcomes.

The ELCAP management system was provided to each participating institution, free of charge. Training for setting up screening programs was provided to each site and is also provided at the international conferences held every 6 months. Training included dual reading by the institution and coordinating center with identification of critical differences as the ELCAP management system provided both the data and images.[37] By the 40th International Conference on Screening held in New York on April 12–13, 2019,[29] the I-ELCAP database has clinical data and images on baseline and repeat rounds on over 82,000 participants throughout the world. The database includes the information on diagnostic work-up, treatment, and follow-up on all participants diagnosed with lung cancer, and also identifies additional diagnoses of metachronous lung cancers during annual follow-up of patients who had been cured of their first lung cancer.

The process of having international meetings every 6 months open to all interested in screening and providing international collaboration for an international health concern has engendered many international collaborations and has enabled an entire family of researchers from many medical, mathematical, engineering, statistics, computing, and other disciplines to work together to continually improve the benefit of LDCT screening for people at risk of lung cancer and other diseases.

The flexibility and utility of the ELCAP cohort design allowed for rapid accumulation of data and thus expansion of knowledge of how to optimize the clinical screening process and the resulting treatment advances.[27,35,41] In fact, considering the latest very low-dose techniques of CT scanners and the findings that can be made on LDCT, LDCT screening should be considered as providing a comprehensive health check of the lungs, heart, and other visualized organs in the LDCT screening,[42] which might prove to be highly cost-effective.[43]

Both the conferences and the management system anticipated the era of “open science, open-source” and have illustrated the power and benefit of an open science approach.[44] The open-science and open-source approach is a formalized approach of the fundamental philosophical underpinning of the Early Diagnosis and Treatment Research Foundation that has provided the ELCAP management system and the international conferences. The ELCAP management system has been made available as an “open source” system for LDCT screening of US veterans who are at high risk of lung cancer and other lung and heart diseases through a grant from the Bristol Myers-Squibb Foundation.[45]

The resulting updated VA-Partnership to increase Access to Lung Screening (VA-PALS)-ELCAP management system is the MUMPS-based software.[46] The ELCAP and updated VA-PALS-ELCAP management systems both use structured forms and radiology reporting, as did the original system, The structured reporting provides the common terminology for characterization of all NCNs, allows for quality assurance assessments, and pooling of data across multi-institutions around the world. In the future, the follow-up recommendations can be automatically generated based on the established screening protocol.

The initial information for each participant is documented in the intake form. Pertinent background information potentially helpful in the interpretation of the LDCT images is collected on the background form, which can be obtained by phone prior to the appointment, at the time the LDCT is performed, or later. As radiologists are responsible for reporting on findings of the entire LDCT, structured reporting of all LDCT findings are documented in the CT evaluation form: NCNs, other lung abnormalities (eg, interstitial lung findings, small airways abnormalities), cardiac abnormalities (eg, coronary artery calcifications and aortic valve calcifications), breast abnormalities (breast density, masses, and other findings), and abdominal abnormalities (eg, liver, adrenal, kidney, esophagus, pancreas, spleen). Input can be done directly or by a dictation template. Upon completion of the CT evaluation form, the written report is automatically generated. The follow-up form is used for documenting whenever the participant returns for a LDCT along with any changes in smoking pattern and diagnosis or hospitalization since the last visit. Biopsy, PET, and treatment forms are available for documenting subsequent diagnostic procedures and interventions.

LDCT screening has been disruptive, as its much earlier identification of lung cancers led to updating of recommendations in related disciplines such as smoking cessation, workup of NCNs, recognition of subtypes of NCNs, and pathology criteria of early lung cancer, staging, and treatment.

As a result of the initial ELCAP study, the investigators recognized that LDCT screening enhanced smoking quit rates and did not encourage resumption of smoking.[47,48] These results led to the strong recommendation that smoking cessation programs should be provided with all LDCT screening programs.[49]

Screening led to ever-increasing understanding of small, early lung cancers as to their appearance, growth, and pathologic characteristics that heretofore had not been well-explored due to insufficient data. The impact of CT scanner technology on tumor growth assessment was demonstrated by the limitations due to the tumor appearance and those due to the quantitative measurements of the scanner itself.[50,51] The importance of measurement error in nodule growth assessment was an important factor leading to the formation of the Quantitative Imaging Biomarker Alliance sponsored by the Radiological Society of North America (RSNA).[52] Hopefully, recognition of these limitations will stimulate development of better quantitative tools so that NCN growth can be determined within weeks rather than months, as this will lead to marked improvement of the efficiency of LDCT and to reducing unnecessary biopsies and/or surgery.

Differences in NCN subtypes, solid, part-solid, and nonsolid, were recognized early in the development of screening protocols.[53-56] The terminology was developed by the ELCAP investigators.[56] After review of the I-ELCAP and NLST databases and then of the world literature,[57-62] recommendations for workup of the different subtypes was developed and incorporated into the I-ELCAP protocol.[19] The usefulness of watchful waiting for nonsolid NCNs and concerns about over-diagnosis was addressed by these publications. The key conclusion was that overtreatment should be avoided and that current surgical recommendations for nonsolid NCNs involve watchful waiting as supported by the revised surgical staging classification of the American Joint Committee on Cancer/International Association for the Study of Lung Cancer (AJCC/IASLC) and address the concerns of over-diagnosis.[63-65] These critical findings allowed for watchful waiting, showing that by annual CT follow-up, cancers identified in nonsolid NCNs remained 100% curable even if they developed a solid component during the year.

Other advances were due to the recognition that the LDCT of the chest provides additional information about the lungs as well as the other visualized organs. The Surgeon General reports had documented the impact of active tobacco smoking and involuntary exposure to tobacco smoke on many other organs, including heart disease, chronic pulmonary diseases, and cancers of other organs.[66-69] Secondhand tobacco smoke exposure was shown to increase the risk of lung cancer, emphysema, and cardiovascular disease,[68] which was also assessed in the context of LDCT screening.[70,71] These results suggest that secondhand tobacco smoke should be included as a risk factor. I-ELCAP was also able to demonstrate that women in the United States were found to be more susceptible to lung cancer,[72,73] but when diagnosed with lung cancer, had a lower risk of dying.[73] The lower risk of women dying of lung cancer was confirmed in the NELSON study.[21]

Pathologic criteria for diagnosis of small, early lung cancer needed updating. The ELCAP investigators had developed a pathology protocol and initiated expert pathology panel reviews in 2000,[74] funded in part by the American Cancer Society. Results of these panel reviews of the pathologic specimens available through I-ELCAP have been published.[25,26,75] These reviews led to the new pathologic classification of adenocarcinoma and to the revised World Health Organization classification.[76-79]

When ELCAP started, the 6th staging classification edition was being used.[24] From the beginning, ELCAP and I-ELCAP also documented the modified tumor size categories as many small cancers were identified: < 10 mm, 11–20 mm, and 21–30 mm. These categories have now been incorporated in the 8th staging edition introduced in 2018.[80,81] Different from the 8th classification, however, in I-ELCAP’s modified staging, multiple adenocarcinomas without lymph node metastases are considered as multiple primary stage I cancers, not as higher stages as currently classified in the 8th classification, based on the evidence resulting from analyses of screening results.[75] The 8th edition of staging does not recognize the impact of nodule consistency (solid, nonsolid, part-solid). I-ELCAP investigators have shown in their recent publications that tumor consistency is both an important diagnostic and prognostic factor that needs to be considered in the tumor, node, and metastasis (TNM) staging, particularly for lung cancers < 30 mm in maximum diameter.[82,83] Based on careful documentation and analyses, investigators have found that clinical staging for tumors 30 mm or smaller should be classified based on the tumor size on CT, as neither lymph node enlargement or F-18-fluorodeoxyglucose (FDG)-PET uptake were helpful in identifying lymph node metastases. These publications called for updates of the 8th classification as data for the smallest size category was limited, despite the impressive contribution of the IASLC staging database.

Breast cancer treatment provides an excellent paradigm of the impact of screening on diagnosis and treatment. Prior to screening, mastectomy with radical lymph node dissection was the standard of care, but currently there are many alternative and less radical treatments. The recommendation of lobar resection has not changed in more than 50 years. Two randomized surgical trials started in 2007,[84,85] but their results are not expected until after 2020. In the meantime, new technologies are being developed (eg, robotic surgery, navigational bronchoscopy, percutaneous ablation) that also need assessment.

Lung cancer will follow a similar paradigm as breast cancer. Published cohort studies are providing outcome results while awaiting the results of randomized trials on the extent of surgical resection.[86-91] Quality of life becomes an important consideration in treatment assessment given the high long-term cure rates of screen-diagnosed lung cancer.[92-95]

Clearly, as recognized in the design of ELCAP, it is the curative treatment of early lung cancer that determines whether screening is effective. As new techniques are continuously being introduced, often with limited data for small lung cancers, the Initiative for Early Lung Cancer Research on Treatment (IELCART) was started using a prospective cohort design as previously used for I-ELCAP.[96] The ELCAP management system has been adapted for treatment considerations and also allows for randomization for future innovative randomized trials within its structure. IELCART is envisioned to be as productive as I-ELCAP has been in producing ongoing evidence.

The low-cost and efficient ELCAP design provided the fundamental clinically relevant information about the benefit of LDCT screening. In the 20 years since our initial ELCAP report in 1999, a vibrant community of LDCT investigators has emerged. Their efforts have ranged from providing the NLST in the United States, the largest randomized screening trial performed, which started in 2002, to the NELSON trial in Europe, which started in 2004.[20,21] The design of the NELSON trial and its use of volumetrics was strongly influenced by extensive discussions of the leading NELSON and ELCAP investigators at I-ELCAP conferences and through personal communications.[22,29] Both the NLST and NELSON trials, when published in 2011 and 2019,[20,21] not only confirmed the ELCAP results reported in 1999, but the NLST also provided a large library of CT images with associated clinical outcomes for research.[97] Important insights of the different screening trial designs and outcome parameters were gained,[98-102] and spinoff screening management optimization approaches were developed.[103-105] Other international trials have reported long-term follow-up of studies with randomized and cohort trial designs across the world that have confirmed the high long-term benefit of CT screening,[28,106-109] thus providing even further assurances as to the robustness of the LDCT screening benefit.

Other screening protocols have emerged since our initial publication in 1999, such as the American College of Radiology Lung Imaging Reporting and Data System (Lung-RADS) and the European Consortium protocols.[103,104] Comparisons of efficiency of these protocols as to the number of workups per diagnosis of lung cancer show that the I-ELCAP is currently the most efficient,[105] but we are excited that each of these efforts provides innovative approaches that benefit all screening participants. As further evidence and technology emerges, it is hoped that protocols will continue to be updated and compared using standardized criteria that measure efficiency and eventually these efforts should lead to further uniform improvements.

 Dr. Reeves receives patent royalties from GE, and is president of D4Vision, Inc. Dr. Yankelevitz owns equity in Accumetra and is on the advisory board of GRAIL. Dr. Henschke and Ms. Yip have no significant financial interest in or other relationship with the manufacturer of any product or provider of any service mentioned in this article.

Innovation is said to be disruptive and the experience in applying low-dose computed tomography (LDCT) to find early lung cancer certainly validated that old bromide. In Dr. Henschke and colleagues’ review, we have the encapsulation of 20 profoundly transformative years in the history of biomedical research. In 1999, lung cancer was a terribly lethal scourge that humbled countless research efforts to improve outcomes. Aside from the courageous interventions of a generation of surgeon generals, a spirited tobacco control community, and a few Japanese epidemiologists, the disease burden of tobacco use was unrelentingly deadly.

The Lancet paper computed tomography (CT) screening report from Dr. Henschke’s team in 1999 was electric and galvanized camps of skeptics and believers alike[1]. This report came at the time of an ongoing and charged reconsideration of screening benefits with breast cancer, as well as prostate cancer. This situation with lung cancer was also colored by memories of initial enthusiasm for chest X-ray screening that did not hold up to more in-depth analysis.

An important, but underappreciated aspect of the New York screening team’s effort was a commitment to continue to pursue research evidence to refine the early lung cancer detection process. This momentum was sustained by a series of workshops that occurred every 6 months for the next 20 years and included advocates, as well as skeptics, from a broad array of medical and scientific disciplines.

Over this time in excess of 300 screening-related manuscripts were published coming from International Early Lung Cancer Action Program (I-ELCAP) collaborative groups in their consortium on aspects from imaging to health policy. Leveraging a remarkably robust informatics capability, the collaboration ultimately accrued over 82,000 screening subjects, including access to over 300,000 thoracic CT scans from participants on five continents. Hundreds of investigators worked with this consortium, including surgical leaders such as Drs. Robert Ginsburg, Nasser Altorki, and Raji Flores. Through coordinated efforts with Drs. William Travis and Adi Gazdar and others from the International Association for the Study of Lung Cancer Pathology Committee, every case of lung cancer detected by ELCAP was reviewed by the pathology panel. That experience led to the changes in international pathology classification and corresponding changes to international staging classifications.

Drs. Harvey Hecht, Matt Cham MD, Joseph Shemesh, and Jagat Narula revealed the dynamic of screen-detected coronary artery disease, while a corresponding revelation of frequent asymptomatic pulmonary parenchymal disease was reported by Javier Zulueta and his Spanish colleagues, principally from the University of Navarra. The list of other new observations is long, from Mary Salvatore team’s findings with interstitial pulmonary fibrosis, to the routine findings of abnormal breast density on screening CTs with Laurie Margolies and colleagues, to recent emerging information about screening for nonalcoholic steatohepatitis (NASH) in the context of low-dose computed tomography screening, with Andrea Branch and colleagues. These activities were thoughtfully acknowledged in Dr. Henshke’s review[2].

An excellent example of the power of this consortial approach was highlighted in a report to the Annals of Internal Medicine on the consequences of selecting the size threshold of detected non-calcified pulmonary nodule relative to the efficiency of a screening management approach. To address this question, Henschke was able to analyze the outcomes of 21, 136 screening cases from a 5-year period of observation to inform the decision as to what threshold would be most appropriate[3] and then further confirm these thresholds using the National Lung Screening Trial Database.[4] Having such robust evidence to refine the efficiency of the screening process reflects how far the I-ELCAP investigators have moved this field.

As Dr. Henschke relates, her group’s work also stimulated competitive and collaborative efforts across the globe with outstanding screening research teams emerging launching the National Lung Screening Trial and the NELSON study, as well as coordinated efforts in Canada, Poland, Spain, Switzerland and many other nations[5-7]. This vibrant experience, while not perfect, did in fact align the contributions of hundreds of talent researchers and advocates to change the world’s perception of and aspirations for lung cancer to the benefit of many.

Dr. Mulshine has no significant financial interest in or other relationship with the manufacturer of any product or provider of any service mentioned in this article.

1) Henschke C, McCauley D, Yankelevitz D, Naidich D, McGuinness G, Miettinen O, Libby D, Pasmantier M, Koizumi J, Altorki N, and Smith J. Early Lung Cancer Action Project: overall design and findings from baseline screening. Lancet. 1999; 354:99-105.

2) Henschke C, Yankelvitz D, Reeves A, Yip R. Evolution of lung cancer screening management. ONCOLOGY. 2019;____ 33-6.

3) Henschke C, Yip R, Yankelevitz DF, Smith JP for the I-ELCAP Investigators. Definition of a positive test result in computed tomography screening for lung cancer: a cohort study. Annals of Internal Medicine. 2013; 158:246-52 PMID: 23420233

4) Yip R, Henschke CI, Yankelevitz DF, Smith JP. CT screening for lung cancer: alternative definitions of positive test result based on the National Lung Screening Trial and International Early Lung Cancer Action Program databases. Radiology 2014; 273: 591-6. Accompanied by a Radiology podcast. PMID: 24955929

5) National Lung Screening Trial Research Team, Aberle DR, Adams AM, Berg CD, Black WC, Clapp JD, et al. (Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011; 365:395-409.

Throughout the last half of the 20th century, despite enormous research efforts, lung cancer remained the most feared as well as lethal cancer across the world. Efforts to update screening management are attempting to change that.

Evolution of Lung Cancer Screening Management

Acute myeloid leukemia (AML) is the most common acute leukemia in adults, with an estimated 21,450 new cases diagnosed annually in the United States.[1] Approximately one third of patients with newly diagnosed AML harbor a mutation in the 

) gene, which encodes a receptor tyrosine kinase involved in proliferation and survival of hematopoietic progenitor cells.[2]Mutations in 

 are associated with poor outcomes with traditional chemotherapy. In particular, the internal tandem duplication 

 mutation is associated with decreased disease-free survival and shorter overall survival (OS) vs patients with wild-type 

Midostaurin, a first-generation multitargeted tyrosine kinase inhibitor, was the first 

 inhibitor to be Food and Drug Administration (FDA) approved for use in AML in the United States.[5] It is indicated for use in the treatment of patients with newly diagnosed 

+ AML in combination with standard induction and consolidation chemotherapy. Its August 2017 approval ended a 17-year drought in new drug approvals for AML and was the first in a wave of targeted agents to be approved for the disease, including the second-generation inhibitor gilteritinib, approved in November 2018 for single-agent treatment of relapsed/refractory 

The availability of these agents represents a paradigm shift in the treatment of 

-mutated AML, and clinicians should be aware of their unique toxicity profiles and prescribing considerations. Of particular interest is the management of drug-drug interactions mediated via cytochrome P450 (CYP) 3A4, as the use of interacting supportive care medications is common during induction and consolidation. This article summarizes available data for the use of CYP3A4 inhibitors with midostaurin.

FDA approval of midostaurin was based on the phase III RATIFY trial, which randomized 717 patients with newly diagnosed 

+ AML to receive midostaurin or placebo (50 mg twice daily on days 8 to 21) in combination with standard induction and consolidation chemotherapy followed by single-agent maintenance therapy (50 mg twice daily on days 1 to 28). The midostaurin arm demonstrated improved OS (median 74.7 months vs 25.6 months; 

 = .009) and event-free survival (EFS) (median 8.2 months vs 3.0 months; 

Adverse events observed in the RATIFY study were comparable between the midostaurin and placebo treatment groups.7 The most frequent grade 3 or higher adverse events were hematologic toxicities, febrile neutropenia, infection, diarrhea, hypokalemia, pain, and increased alanine aminotransferase. Rates of grade ≥3 anemia (92.7% vs 87.8%, 

 = .008) were higher with midostaurin versus placebo. While grade ≥3 nausea occurred more frequently in the placebo group (5.6% vs 9.6%, 

This article summarizes available data for the use of CYP3A4 inhibitors with midostaurin for acute myeloid leukemia.

ONCOLOGY Vol 33 No 7