- Email: [email protected]
A comprehensive review of exceptional responders to anticancer drugs in the biomedical literature
European Journal of Cancer 101 (2018) 143e151
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
Original Research
A comprehensive review of exceptional responders to anticancer drugs in the biomedical literature Go Nishikawa d,1, Jia Luo d,1, Vinay Prasad a,b,c,* a
Division of Hematology Oncology, Knight Cancer Institute, Oregon Health & Science University, USA Department of Public Health and Preventive Medicine, Oregon Health & Science University, USA c Senior Scholar in the Center for Health Care Ethics, Oregon Health & Science University, USA d Department of Medicine, Oregon Health & Science University, USA b
Received 11 June 2018; accepted 13 June 2018
KEYWORDS Exceptional responders; Super responders; Dramatic response; Cancer outliers; n of 1 medicine
Abstract Importance: There is broad interest in patients who have exceptional responses to cancer drugs and only one prior systematic review of published case reports. Objective: To systematically summarise case reports of patients who have had a super response to a cancer drug and determine patterns and completeness of documentation. Design, setting and participants: This is a systematic search of biomedical literature and hand searches of seven prominent oncology journals. All cases were adults with cancer in the metastatic or unresectable setting and who were classified as having a remarkable response to a drug therapy were included. Our study was conducted between January 2017 and April 2017. Main outcomes and measures: We extracted patient and tumour characteristics, characteristics surrounding the exceptional response and pertinent oncologic history. Additionally, we noted authors’ financial conflicts of interest, if given. Results: We found 180 cases of exceptional response. The most common malignancy was breast cancer (28/180), and the most common class of drug therapy used was targeted therapies (161/274). The median year of publication was 2015. Depth of response included 33% (61/ 180) complete response, 55% (100/180) partial response and 7.7% (14/180) stable disease. The median length of response to the drug was 14 months. Fifty-nine percent (107/180) reported a driver mutation, and only 52% (56/107) reported two or more correlative studies. Only 14% (26/180) reported a denominator or total number treated to observe this response. Of cases that reported duration of response to a previous drug given in the unresectable setting, 49% demonstrated a progression-free survival ratio of exceptional response to prior line of 1.3 or greater (Von Hoff ratio).
* Corresponding author: Oregon Health & Science University, 3181 SW Sam Jackson Park Road, Portland, OR 97239, USA. Fax: þ503 494 3257. E-mail address: [email protected] (V. Prasad). 1 These two authors contributed equally to this work. https://doi.org/10.1016/j.ejca.2018.06.010 0959-8049/ª 2018 Elsevier Ltd. All rights reserved.
144
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
Conclusions and relevance: Our survey found that exceptional responders tended to have solid organ malignancies and most received targeted therapies. Many of our cases of exceptional responders were missing important information about the response and key oncologic history. Standardised criteria should be used to maximise usefulness of these case reports in the future. ª 2018 Elsevier Ltd. All rights reserved.
Key points: Question e What characteristics are reported in case reports of exceptional responders to cancer drugs in the literature and how complete are the reports? Findings e In our review, we found 180 cases of exceptional responders to cancer drugs, mostly patients with solid tumours responding to targeted agents. Many reports did not contain enough information to fully interpret the depth of response, and some of the cases may not be true super responders. Meaning e In case reports of exceptional responders, to maximise clinical benefit, we propose a full set of criteria to report, including patient and tumour characteristics, characteristics of the super drug and complete description of prior and non-drug treatments.
1. Introduction There is widespread enthusiasm for precision cancer medicine, including the use of whole-exome tumour genomic testing to select targeted therapies for patients and to better understand why some patients may respond dramatically to treatment, while others do not [13]. In 2015, the National Cancer Institute (NCI) launched the Exceptional Responders Initiative to molecularly characterise patient tumours that dramatically responded to the [3e5]. As of 06 July 2017, the NCI Exceptional Responders Initiative reports that it has enrolled 40 patients, although it has not yet provided an update of its findings [6]. While we await results of prospective studies, published case reports may serve as an existing way to characterise atypical disease courses in the literature, including super responders to cancer drugs. In fact, one case report of a super responder to everolimus is widely cited as an example of the importance of studying exceptional responders [6,7]. Therefore, case reports may serve as a complementary data set to what is being collected by the NCI. For this reason, our goal was to collect and summarise all existing reports of patients who have had extraordinary responses to anticancer drugs in the medical literature. We sought to examine factors that
help readers assess whether the benefits seen in these cases were attributable to the cancer drug and the rate at which these factors were reported. This builds on our prior summary of exceptional responders [8]. Additionally, we noted conflicts of interest among authors of these case reports. 2. Methods We set out to review case reports where authors claim a drug resulted in an exceptional or super clinical outcomes. 2.1. Search criteria and study inclusion There were four components to our search strategy. First, we included all cases described in a prior systematic review of exceptional and super responders [8]. The prior report was limited to 32 cases. Second, we expanded this by searching MEDLINE (PubMed), Scopus and Google Scholar for [‘exceptional response’], [‘exceptional responder’], [‘outstanding response’], [‘outstanding responder’], [‘super response’], [‘super responder’], [‘miracle response’] and [‘miracle responder’] without further limits on the search on 01 April 2017. Third, we hand searched several prominent journals from January 2010 to April 2017, namely, Annals of Oncology, Clinical Cancer Research, JAMA Oncology, Journal of Clinical Oncology (JCO), JCO Precision Oncology, Journal of the National Cancer Institute and Lancet Oncology. This was performed between 05 April 2017 and 15 April 2017. Fourth, for all articles identified through the previously mentioned three strategies, we reviewed all references manually. We sought case reports of metastatic, unresectable or relapsed cancer where the prolonged response was attributed to the drug the patient was receiving in the opinion of the authors. Articles were screened initially based on title, then abstract, then full text. We included reports of chemotherapy, targeted therapy, immunotherapy and other drugs or biologics. We included cases using multiple treatment modalities. Because our focus was on drugs, we excluded studies solely of surgery or radiation therapy. Case reports of remarkable anticancer drugs in the neoadjuvant setting, case reports of patients younger than 18 years and case reports of patients whose cancers spontaneously resolved without
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
any treatment were excluded. We excluded cases only reported as abstract. We excluded two articles that were not in English. Two reviewers independently performed the searches. Inclusion and exclusion criteria were agreed on among all authors, and a consensus was reached about equivocal cases before data analysis. 2.2. Data collection We collected the following information from each case report: histologic type of cancer, age of the patient at time of treatment, publication year, drug leading to the impressive response, best response (stable disease [SD], partial response [PR] or complete response [CR]), duration of best response or duration of SD, the number of prior lines of therapy, the duration of prior responses and finally, the number of patients treated similarly to observe this response. We characterised the final status of the patient as alive without disease progression (alive), alive with progressive disease (progression) or deceased. Cases that were unclear regarding the patient’s final status were characterised as alive without progression. Additionally, we documented whether a driver mutation resulted in the response, and, if so, the number of correlative studies performed to demonstrate this. Correlative studies were further characterised by type, such as next-generation sequencing, immunohistochemistry, etc. Additional, consolidative and confounding therapies patients received were also collected, which included surgery, radiation therapy and stem cell transplant. Two variables were either reported directly from the case or calculated based on reported dates in the case, if present: the longest duration of response to a prior therapy and the duration of response from the metastatic/unresectable diagnosis to initiation of the drug leading to the remarkable response. Finally, the conflict of interest (COI) section was reviewed to note if any of the authors had a financial relationship to the drug the patient received or any reported company of the correlative test(s) used. 2.3. Statistical analysis Descriptive statistics are reported where appropriate. The swimmer’s plot was made in python using matplotlib. We defined progression-free survival (PFS) of the exceptional drug (PFSex) as the duration of response to the drug characterised as an exceptional response and PFSprev as the single longest duration of response to a prior drug since metastatic or unresectable disease. The PFSex/PFSprev graph and pie graph were made in Microsoft Excel (14.4.7). This study of published case reports did not involve the use of human subjects and did not require IRB approval. Our study was conducted between January 2017 and October 2017.
145
3. Results We identified 10148 articles. A CONSORT diagram is available in the Supplement (Supplementary Table 1). After review, 180 case reports of a remarkable response to an anticancer drug or drugs from 134 articles were identified. A full description of details of all cases appears in the Supplement (Supplementary table 2). 3.1. General characteristics of case reports All cases were stage IV or metastatic/recurrent or unresectable tumours. The median age was 56 years old (range 18e89). The most common type of cancer was breast cancer (28/180, 15.5%), followed by non-smallcell lung cancer (NSCLC) (18/180, 10%) and melanoma (17/180, 9.4%). The median year of publication was 2015 (range 2007e2017), and the median year of treatment initiation was 2010 (range 1998e2016). 3.2. Characteristics of the disease and drug credited with the super response Among 180 cases, 274 drugs led to the exceptional or super response, either as monotherapy or in combination. Targeted therapies accounted for 59% (161/274) of drugs used, 32% (87/274) were cytotoxic therapies, 6.9% (19/274) immunotherapies and 2% (6/274) were classified as other drugs. Of all drugs, 93% (255/274) were Food and Drug Administration (FDA) approved, and 7% (19/ 274) were experimental agents. Among the types of targeted therapies, the most common subclass was small molecule kinase inhibitors (KIs; 48%, 78/161), followed by monoclonal antibodies (25%, 41/161) and mammalian target of rapamycin inhibitors (15%, 24/161). Among the KIs, the most common targets were serine/threonine-protein kinase B-Raf (BRAF) (30%, 23/78), vascular endothelial growth factor (VEGF) (30%, 23/78) and epidermal growth factor receptor (EGFR) (25%, 19/78). We examined the depth of response to the exceptional drug. CR was reported in 33% (61/180) of the cases. A PR was reported in 55% (100/180) and SD in 7.7% (14/180). Among the cases, 17.7% (32/180) did not formally report the type of response to the super drug and had to be inferred based on the wording of the text. The median length of response was 14 months (n Z 169 reported). At time of publication, 74% (133/180) of the patients in the case reports were alive or no final status was reported, 20% (36/180) had progressive disease and 6.1% (11/180) died. In total, 59% (107/180) of case reports reported a putative driver mutation. Of those 107 cases, 43% (46/107) reported one additional correlative study and 52% (56/ 107) reported two or more additional correlative studies. The median of number of prior lines of therapy was 1 (range Z 0 to 10, n Z 156 reported). Fifty-seven percent
146
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
Table 1 Proportion of characteristics that were reported in case reports of exceptional responders. Characteristic
Fraction (%)
Histologic type and cancer Correlative study Length of documented response Age of patient Number of previous lines of therapy Type of response were reported Best duration of response to a previous single line of therapy Duration since metastasis Year of treatment Number of similar patients treated to observe the response
180/180 (100%) 102/107a (95.3%) 169/180 (93.9%) 165/180 (91.7%) 156/180 (86.7%) 148/180 (822%) 80/133b (60.2%) 71/133b (53.3%) 88/180 (48.9%) 26/180 (14.4%)
Duration since metastasis Z Duration of response from diagnosis of metastatic disease to prior to the start of the exceptional drug. a Of the 180 cases, 107 cases reported at least one driver mutation. b Of the 180 cases, 133 cases used two or more lines of therapy.
of patients (102/180) also had undergone at least one surgical procedure, stem cell transplant or radiation, and 18% of patients (32/180) underwent a procedure directed at treatment of oligometastatic disease. The median best duration of response to a prior therapy was 8 months (range Z 1 to 156) among the 80 cases that reported this. Only 14% (26/180) of cases reported how many others were treated with the same drug to observe this response, and among those the median was 160 patients (range Z 5 to 210). We list the fraction of cases that report pertinent information in Table 1. We note 86% (154/180) of cases did not report the number of patients treated similarly, that is, the number of patients treated with the drug on a protocol or if, off protocol, the number of patients treated with the use of that genetic test (e.g. Foundation Medicine or in-house next generation sequencing (NGS) panel). In other words, some estimate of the denominator from which the case arose. 3.3. Comparing super responses with prior responses For the 106 (of 180) reports that provided duration of response to the exceptional drug as well as past therapies if existed, we constructed a swimmers plot to examine response to therapy since metastatic or unresectable disease (Fig. 1). Of these cases, 32% (34/106) received the exceptional drug as initial therapy. Multiple cases show a substantial duration of response to prior therapies before receipt of the exceptional drug (Fig. 1). When reported, we compared the ratio of the best duration of response to a prior therapy and the duration of response to the present therapy. There were 79 cases that reported this information. Fig. 2 depicts this. Of these cases, 56% (45/79) that showed a PFS to the exceptional drug was greater than a prior PFS. When using the more stringent Von Hoff criteria (which asks for the PFSex/PFSprev 1.3), 49% (39/79) cases demonstrated this characteristic [9].
3.4. Potential financial conflict of interests In 22% (40/180), the authors declared a COI to the drug that was reported to lead to the exceptional response. In 15% (27/180), the authors declared a conflict with the test that was used to characterise an exceptional response. In 57% (n Z 103) of cases, the authors reported a COI to the biomedical industry in general. We noted that case reports with a financial conflict to the test (often a genomic test) were more likely to omit data (missing elements present in no COI [median Z 2, 25th percentile Z 2, 75th percentile Z 3], COI to the test company [median Z 4, 25th percentile Z 2, 75th percentile Z 5], COI to the drug company [median Z 2, 25th percentile Z 1, 75th percentile Z 4] and COI not related to the test or drug company [median Z 2.5, 25th percentile Z 1, 75th percentile Z 3]). Using analysis of variance with a Bonferroni correction for multiple comparisons, we found that articles with a declared COI to the test company had a significantly greater number of missing observations than articles with no declared COI (p-value < .001), COI to the drug company (p-value Z .032) and with COI not related to the test or drug company (p-value Z .019). All other comparisons were not significantly different. 4. Discussion We examined characteristics of case reports of super responders, noting key omissions. Our analysis of the published biomedical literature complements the results of forthcoming prospective studies, such as the NCI’s Exceptional Responders Initiative, and provides guidance regarding a standardised set of reporting criteria of these cases. 4.1. Incomplete reporting To draw conclusions from reports, cases should be well documented. We found that many studies
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
147
Fig. 1. Swimmers plot of cases that reported response since metastatic disease. We identified 106 cases that reported duration of therapy since metastatic or unresectable disease. The graph shows previous total response, if the patient received prior therapy in gray. The colour bars represent response to the super drug per the authors of the article. Symbols represent the final status of the patient. Ambiguous wording of cases were mutually settled on by all the authors collectively. CML, chronic myeloid leukemia; CR, complete response; MDS, myelodysplastic syndrome; PR, partial response, SD, stable disease; NSCLC, nonesmall-cell lung cancer; SCC, squamous cell carcinoma.
148
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
reporting exceptional responders were insufficiently informative and were missing information that could help inform whether the case is actually a super response to a drug. Of all the cases, 51% (92/180) left out the year of treatment. This is important in providing the historical context of when the therapy was started and informs readers what other therapies were standard of care at that time. We noted that duration of response was not reported in 47% (62/133) of cases where patients were treated with multiple lines after metastatic disease, and 40% (53/133) of cases were missing best prior response. These values are useful to report and to compare the exceptional drug with prior drugs to note its relative effectiveness. The
total number of patients treated to observe the exceptional case was reported in a small fraction (14%, 26/180) of cases. This is important in that the denominator can help inform the likelihood an individual would respond to this drug and also shed light on the potential toxicity of this therapy among those who do not respond. Additionally, some cases left out the depth of response (e.g. CR, PR, SD) of the exceptional drug (18%, 32/180), which is critical for evaluating whether a response was present. The lack of complete reporting makes it challenging for clinicians to use this information to guide clinical decision-making and researchers to generate hypotheses for further studying patients who fall into this category.
Fig. 2. A number of cases had a better best prior response compared with response to the exceptional drug. Seventy-nine cases reported both the duration of response to prior lines of therapy and the duration of response to the super drug. This graph represents the PFS of the exceptional drug (PFSex) as compared to the PFS of the best prior response (PFSprev) since metastatic/unresectable disease. The dashed line represents PFSex/PFSprev Z 1. The solid line represents PFSex/PFSprev 1.3. This corresponded to 56% and 49% of the cases, respectively. PFS, progression-free survival.
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
149
Fig. 3. Authors’ financial conflicts of interest. This pie chart notes declared conflicts of interest of any of the authors of the 180 cases we characterised.
4.2. Many cases of exceptional responders had confounding elements
4.3. Exceptional responders are likely a very small set of patients with cancer
For the cases that reported duration of response, including the super response, it was sometimes not clear that the dramatic response was significant in the context of the patient’s disease course. We illustrate this with a swimmer’s plot of all the cases that reported response durations (Fig. 1). The most compelling cases of patients with an exceptional response in Fig. 1 are likely those with a shorter segment for previous total response (in gray) followed by a longer CR (in green), reflecting a true atypical response. On the other hand, some of these cases may represent indolent diseasedi.e. consistently long responses irrespective of the drug. This can be the case even for known aggressive disease states; for example, a Netherlands Cancer Registry retrospective analysis found that the 5-year survival of pathologically verified pancreatic adenocarcinoma was 1.7% [10]. Cases such as those with a prolonged previous total response (in gray) may reflect indolent or pansensitive tumour biology. For example, one notable case we found is a patient with a lung adenocarcinoma who underwent six previous lines of therapy for 65 months before receiving the drugs that led to the exceptional response [11]. Additionally, we noted that of cases that had response durations for at least one other therapy, 44% (34/79) had a longer response to the prior drug compared with the reported super drug. This percentage increased to 51% (40/79) if the more stringent Von Hoff criteria of 1.3 times the prior response were used. Furthermore, we found that a substantial number 32/ 180 (18%) of cases underwent treatment directed at eradicating oligometastatic disease, which can serve to confound exceptional responses seen.
We identified 180 case reports of super response. It is worth putting these roughly in contrast against the number of patients who have undergone next-generation sequencing or comparable commercial testing. At least 120,000 cancer patients have had their tumours profiled through Foundation Medicine, Guardian 360 and the GENIE consortium [12e14]. Although it must be noted, many cases of super response may not be reported in the literature. 4.4. Cases reported few correlative studies to support selection of targeted super drug Of the 274 drugs used as monotherapy or in combination, 161 were targeted therapies. Many of the cases (59%, 107/180) report a driver mutation, with most of these cases reporting the type of genomic analysis used (95%, 102/107). We noted that 43% (46/107) of cases reported just one correlative study. Recently multiple prominent investigations in solid tumours such as lung and renal cell have identified cancers as having significant heterogeneity [15,16]. This makes one question whether simply performing one correlative study will fairly represent a patient’s disease. Additionally, researchers have argued that a prerequisite to understanding the pathophysiology of an individual’s cancer requires -omic, epigenetic and functional studies from multiple sources (tumour, blood, metastatic site) [17,18]. 4.5. Reporting of exceptional responders moving forward We found that 16.1% of our cases (29/180) had duration of response to the exceptional drug of fewer than 6
150
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
Table 2 Recommended characteristics to report in case reports claiming remarkable response to drug. Characteristic
Information that should be reported
Tumor characteristics
Disease Stage
Patient characteristics
Age of patient Last known final status of patient
Characteristics of exceptional drug
Year of treatment Drug used
Cancer type and relevant histology Stage the patient was diagnosed with cancer as well as the stage the patient started the exceptional drug Age the patient was treated with the drug Language clearly stating whether the patient remains alive, developed progressive disease, or died at the time of publication Year the patient started the drug Drug that led to remarkable the response along with all prior lines of therapy and non-drug therapies. Should discuss if this was part of clinical trial. Based on RECIST criteria for solid tumors or disease specific response criteria for hematologic malignancies From time of initiation of drug to disease progression on exceptional drug or status at publication Should include whether the study was conducted as clinical trial or if it is a report of a single exceptional case Should mention how likely the mutation is attributable to the super response, and whether the mutation is new or a known mutation Include test performed, sample tested (blood, primary tumor, metastatectomy etc). Should include test company or where the tests were performed Mention if previous lines include neoadjuvant or adjuvant, or if enrolled in experimental trial Longest response to a previous line of therapy
Type of response
Studies to identify potential driver mutation
Length of documented response to exceptional drug Number of similar patients treated to observe the response Presence of driver mutation
Correlative studies
Prior drug treatment
Other treatment modalities used
Number of previous lines of therapy Best duration of response to a previous single line of therapy Duration of response since metastasis to prior to start of exceptional drug Previous surgery, transplant or radiation Presence of metastectomy or oligometastatic radiotherapy
months. Because the NCI’s Exceptional Responder’s initiative defines that an exceptional responder case must have either a CR or a PR with duration of at least 6 months, those cases would not meet this criterion [19]. Given the importance of complete reporting, we have devised an essential list of items that authors should consider reporting when writing a case report about a patient who has had an exceptional response to a cancer drug (Table 2).This information includes detailed characteristics and history of the patient’s disease course as well as correlative studies. 4.6. Study limitations Our study has several limitations. Although we used several, disparate search strategies, it is possible we missed some articles. Furthermore, not all cases of exceptional responders may be published in the literature. However, we screened over 10,000 publications, which represents the largest effort to date. Second, our reporting criteria were prespecified, but reasonable people may disagree that all of these elements are necessary or alternatively may wish to include aspects that we did not. For this reason, we believe our analysis is merely the starting point for a broader discussion about the core elements that should be reported in all
Should include sites, frequency, and timing of intervention Should include sites, frequency, and timing of intervention
cases of super response. Table 2 is our suggestion for this starting point. 5. Conclusion We found 180 cases of super response to a cancer drug. Many were missing key information about the nature of the response and pertinent past oncologic history. When the current duration of response could be compared with the best prior duration, less than half had a ratio greater than 1.3. To maximise the usefulness of these case reports in the future, standardised reporting criteria are needed. Contributions All authors designed, conducted and drafted the article. Conflict of interest statement Vinay Prasad reports receiving royalties from his book Ending Medical Reversal; that his work is funded by the Laura and John Arnold Foundation; that he has received honoraria for Grand Rounds/lectures from several universities, medical centers, and professional societies and payments for contributions to Medscape.
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
The other authors declare that they have no conflict of interest to disclose. Authors’ declaration The authors confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. They further confirm that the order of authors listed in the manuscript has been approved by all of them. They confirm that they have given due consideration to the protection of intellectual property associated with this work and that there are no impediments to publication, including the timing of publication, with respect to intellectual property. In so doing they confirm that they have followed the regulations of their institutions concerning intellectual property. They understand that the Corresponding Author is the sole contact for the editorial process (including Editorial Manager and direct communications with the office). He is responsible for communicating with the other authors about progress, submissions of revisions and final approval of proofs.
Acknowledgements Daniel Lecoanet for figure illustration assistance. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.ejca.2018.06.010.
References [1] Hyman DM, Taylor BS, Baselga J. Implementing genome-driven oncology. Cell 2017;168(4):584e99. [2] Takebe N, McShane L, Conley B. Biomarkers: exceptional responders-discovering predictive biomarkers. Nat Rev Clin Oncol 2015;12(3):132e4. [3] Marx V. Cancer: a most exceptional response. Nature 2015; 520(7547):389e93.
151
[4] Printz C. NCI launches exceptional responders initiative: researchers will attempt to identify why some patients respond to treatment so much better than others. Cancer 2015;121(6):803e4. [5] Conley B, Ivy S, Tricoli J, et al. The NCI exceptional responders initiatives: initial feasibility result [Abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research75; 2015 (15 Suppl):Abstract nr 612. [6] Cancer researchers report progress in studying exceptional responders. National cancer Institute NIH; 2017. https://www. cancer.gov/news-events/cancer-currents-blog/2017/exceptionalresponders-progress. [Accessed 26 October 2017]. [7] Iyer G, Hanrahan AJ, Milowsky MI, et al. Genome sequencing identifies a basis for everolimus sensitivity. Science (New York, NY). 2012;338(6104):221. [8] Prasad V, Vandross A. Characteristics of exceptional or super responders to cancer drugs. Mayo Clin Proc 2015;90(12): 1639e49. [9] Von Hoff DD, Stephenson Jr JJ, Rosen P, et al. Pilot study using molecular profiling of patients’ tumors to find potential targets and select treatments for their refractory cancers. J Clin Oncol: Off J Am Soc Clin Oncol 2010;28(33):4877e83. [10] Zijlstra M, Bernards N, de Hingh IH, et al. Does long-term survival exist in pancreatic adenocarcinoma? Acta Oncol (Stockholm, Sweden). 2016;55(3):259e64. [11] Chiari R, Duranti S, Ludovini V, et al. Long-term response to gefitinib and crizotinib in lung adenocarcinoma harboring both epidermal growth factor receptor mutation and EML4-ALK fusion gene. J Clin Oncol: Off J Am Soc Clin Oncol 2014;32(9): e30e2. [12] Foundation Medicine Announces. 2016 second quarter results and recent highlights [press release]. Foundation medicine [Online]. Foundation Medicine, Inc; October 26, 2017 2016. [13] Teichert E. Innovations: startup’s liquid biopsies match cancer patients with trials [Online] Mod Healthc 2016. http://www. modernhealthcare.com/article/20160820/MAGAZINE/ 308209978. [Accessed 26 October 2017]. [14] Consortium APG. AACR project GENIE: powering precision medicine through an international consortium. Canc Discov 2017; 7(8):818e31. [15] Jamal-Hanjani M, Wilson GA, McGranahan N, et al. Tracking the evolution of non-small-cell lung cancer. N Engl J Med 2017; 376(22):2109e21. [16] Gerlinger M, Rowan AJ, Horswell S, et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med 2012;366(10):883e92. [17] Letai A. Functional precision cancer medicine-moving beyond pure genomics. Nat Med 2017;23(9):1028e35. [18] Alizadeh AA, Aranda V, Bardelli A, et al. Toward understanding and exploiting tumor heterogeneity. Nat Med 2015;21(8):846e53. [19] Exceptional responders initiative: questions and answers. National Cancer Institute NIH; 2017. https://www.cancer.gov/aboutcancer/treatment/research/exceptional-responders-initiative-qa. [Accessed 26 October 2017].
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
Original Research
A comprehensive review of exceptional responders to anticancer drugs in the biomedical literature Go Nishikawa d,1, Jia Luo d,1, Vinay Prasad a,b,c,* a
Division of Hematology Oncology, Knight Cancer Institute, Oregon Health & Science University, USA Department of Public Health and Preventive Medicine, Oregon Health & Science University, USA c Senior Scholar in the Center for Health Care Ethics, Oregon Health & Science University, USA d Department of Medicine, Oregon Health & Science University, USA b
Received 11 June 2018; accepted 13 June 2018
KEYWORDS Exceptional responders; Super responders; Dramatic response; Cancer outliers; n of 1 medicine
Abstract Importance: There is broad interest in patients who have exceptional responses to cancer drugs and only one prior systematic review of published case reports. Objective: To systematically summarise case reports of patients who have had a super response to a cancer drug and determine patterns and completeness of documentation. Design, setting and participants: This is a systematic search of biomedical literature and hand searches of seven prominent oncology journals. All cases were adults with cancer in the metastatic or unresectable setting and who were classified as having a remarkable response to a drug therapy were included. Our study was conducted between January 2017 and April 2017. Main outcomes and measures: We extracted patient and tumour characteristics, characteristics surrounding the exceptional response and pertinent oncologic history. Additionally, we noted authors’ financial conflicts of interest, if given. Results: We found 180 cases of exceptional response. The most common malignancy was breast cancer (28/180), and the most common class of drug therapy used was targeted therapies (161/274). The median year of publication was 2015. Depth of response included 33% (61/ 180) complete response, 55% (100/180) partial response and 7.7% (14/180) stable disease. The median length of response to the drug was 14 months. Fifty-nine percent (107/180) reported a driver mutation, and only 52% (56/107) reported two or more correlative studies. Only 14% (26/180) reported a denominator or total number treated to observe this response. Of cases that reported duration of response to a previous drug given in the unresectable setting, 49% demonstrated a progression-free survival ratio of exceptional response to prior line of 1.3 or greater (Von Hoff ratio).
* Corresponding author: Oregon Health & Science University, 3181 SW Sam Jackson Park Road, Portland, OR 97239, USA. Fax: þ503 494 3257. E-mail address: [email protected] (V. Prasad). 1 These two authors contributed equally to this work. https://doi.org/10.1016/j.ejca.2018.06.010 0959-8049/ª 2018 Elsevier Ltd. All rights reserved.
144
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
Conclusions and relevance: Our survey found that exceptional responders tended to have solid organ malignancies and most received targeted therapies. Many of our cases of exceptional responders were missing important information about the response and key oncologic history. Standardised criteria should be used to maximise usefulness of these case reports in the future. ª 2018 Elsevier Ltd. All rights reserved.
Key points: Question e What characteristics are reported in case reports of exceptional responders to cancer drugs in the literature and how complete are the reports? Findings e In our review, we found 180 cases of exceptional responders to cancer drugs, mostly patients with solid tumours responding to targeted agents. Many reports did not contain enough information to fully interpret the depth of response, and some of the cases may not be true super responders. Meaning e In case reports of exceptional responders, to maximise clinical benefit, we propose a full set of criteria to report, including patient and tumour characteristics, characteristics of the super drug and complete description of prior and non-drug treatments.
1. Introduction There is widespread enthusiasm for precision cancer medicine, including the use of whole-exome tumour genomic testing to select targeted therapies for patients and to better understand why some patients may respond dramatically to treatment, while others do not [13]. In 2015, the National Cancer Institute (NCI) launched the Exceptional Responders Initiative to molecularly characterise patient tumours that dramatically responded to the [3e5]. As of 06 July 2017, the NCI Exceptional Responders Initiative reports that it has enrolled 40 patients, although it has not yet provided an update of its findings [6]. While we await results of prospective studies, published case reports may serve as an existing way to characterise atypical disease courses in the literature, including super responders to cancer drugs. In fact, one case report of a super responder to everolimus is widely cited as an example of the importance of studying exceptional responders [6,7]. Therefore, case reports may serve as a complementary data set to what is being collected by the NCI. For this reason, our goal was to collect and summarise all existing reports of patients who have had extraordinary responses to anticancer drugs in the medical literature. We sought to examine factors that
help readers assess whether the benefits seen in these cases were attributable to the cancer drug and the rate at which these factors were reported. This builds on our prior summary of exceptional responders [8]. Additionally, we noted conflicts of interest among authors of these case reports. 2. Methods We set out to review case reports where authors claim a drug resulted in an exceptional or super clinical outcomes. 2.1. Search criteria and study inclusion There were four components to our search strategy. First, we included all cases described in a prior systematic review of exceptional and super responders [8]. The prior report was limited to 32 cases. Second, we expanded this by searching MEDLINE (PubMed), Scopus and Google Scholar for [‘exceptional response’], [‘exceptional responder’], [‘outstanding response’], [‘outstanding responder’], [‘super response’], [‘super responder’], [‘miracle response’] and [‘miracle responder’] without further limits on the search on 01 April 2017. Third, we hand searched several prominent journals from January 2010 to April 2017, namely, Annals of Oncology, Clinical Cancer Research, JAMA Oncology, Journal of Clinical Oncology (JCO), JCO Precision Oncology, Journal of the National Cancer Institute and Lancet Oncology. This was performed between 05 April 2017 and 15 April 2017. Fourth, for all articles identified through the previously mentioned three strategies, we reviewed all references manually. We sought case reports of metastatic, unresectable or relapsed cancer where the prolonged response was attributed to the drug the patient was receiving in the opinion of the authors. Articles were screened initially based on title, then abstract, then full text. We included reports of chemotherapy, targeted therapy, immunotherapy and other drugs or biologics. We included cases using multiple treatment modalities. Because our focus was on drugs, we excluded studies solely of surgery or radiation therapy. Case reports of remarkable anticancer drugs in the neoadjuvant setting, case reports of patients younger than 18 years and case reports of patients whose cancers spontaneously resolved without
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
any treatment were excluded. We excluded cases only reported as abstract. We excluded two articles that were not in English. Two reviewers independently performed the searches. Inclusion and exclusion criteria were agreed on among all authors, and a consensus was reached about equivocal cases before data analysis. 2.2. Data collection We collected the following information from each case report: histologic type of cancer, age of the patient at time of treatment, publication year, drug leading to the impressive response, best response (stable disease [SD], partial response [PR] or complete response [CR]), duration of best response or duration of SD, the number of prior lines of therapy, the duration of prior responses and finally, the number of patients treated similarly to observe this response. We characterised the final status of the patient as alive without disease progression (alive), alive with progressive disease (progression) or deceased. Cases that were unclear regarding the patient’s final status were characterised as alive without progression. Additionally, we documented whether a driver mutation resulted in the response, and, if so, the number of correlative studies performed to demonstrate this. Correlative studies were further characterised by type, such as next-generation sequencing, immunohistochemistry, etc. Additional, consolidative and confounding therapies patients received were also collected, which included surgery, radiation therapy and stem cell transplant. Two variables were either reported directly from the case or calculated based on reported dates in the case, if present: the longest duration of response to a prior therapy and the duration of response from the metastatic/unresectable diagnosis to initiation of the drug leading to the remarkable response. Finally, the conflict of interest (COI) section was reviewed to note if any of the authors had a financial relationship to the drug the patient received or any reported company of the correlative test(s) used. 2.3. Statistical analysis Descriptive statistics are reported where appropriate. The swimmer’s plot was made in python using matplotlib. We defined progression-free survival (PFS) of the exceptional drug (PFSex) as the duration of response to the drug characterised as an exceptional response and PFSprev as the single longest duration of response to a prior drug since metastatic or unresectable disease. The PFSex/PFSprev graph and pie graph were made in Microsoft Excel (14.4.7). This study of published case reports did not involve the use of human subjects and did not require IRB approval. Our study was conducted between January 2017 and October 2017.
145
3. Results We identified 10148 articles. A CONSORT diagram is available in the Supplement (Supplementary Table 1). After review, 180 case reports of a remarkable response to an anticancer drug or drugs from 134 articles were identified. A full description of details of all cases appears in the Supplement (Supplementary table 2). 3.1. General characteristics of case reports All cases were stage IV or metastatic/recurrent or unresectable tumours. The median age was 56 years old (range 18e89). The most common type of cancer was breast cancer (28/180, 15.5%), followed by non-smallcell lung cancer (NSCLC) (18/180, 10%) and melanoma (17/180, 9.4%). The median year of publication was 2015 (range 2007e2017), and the median year of treatment initiation was 2010 (range 1998e2016). 3.2. Characteristics of the disease and drug credited with the super response Among 180 cases, 274 drugs led to the exceptional or super response, either as monotherapy or in combination. Targeted therapies accounted for 59% (161/274) of drugs used, 32% (87/274) were cytotoxic therapies, 6.9% (19/274) immunotherapies and 2% (6/274) were classified as other drugs. Of all drugs, 93% (255/274) were Food and Drug Administration (FDA) approved, and 7% (19/ 274) were experimental agents. Among the types of targeted therapies, the most common subclass was small molecule kinase inhibitors (KIs; 48%, 78/161), followed by monoclonal antibodies (25%, 41/161) and mammalian target of rapamycin inhibitors (15%, 24/161). Among the KIs, the most common targets were serine/threonine-protein kinase B-Raf (BRAF) (30%, 23/78), vascular endothelial growth factor (VEGF) (30%, 23/78) and epidermal growth factor receptor (EGFR) (25%, 19/78). We examined the depth of response to the exceptional drug. CR was reported in 33% (61/180) of the cases. A PR was reported in 55% (100/180) and SD in 7.7% (14/180). Among the cases, 17.7% (32/180) did not formally report the type of response to the super drug and had to be inferred based on the wording of the text. The median length of response was 14 months (n Z 169 reported). At time of publication, 74% (133/180) of the patients in the case reports were alive or no final status was reported, 20% (36/180) had progressive disease and 6.1% (11/180) died. In total, 59% (107/180) of case reports reported a putative driver mutation. Of those 107 cases, 43% (46/107) reported one additional correlative study and 52% (56/ 107) reported two or more additional correlative studies. The median of number of prior lines of therapy was 1 (range Z 0 to 10, n Z 156 reported). Fifty-seven percent
146
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
Table 1 Proportion of characteristics that were reported in case reports of exceptional responders. Characteristic
Fraction (%)
Histologic type and cancer Correlative study Length of documented response Age of patient Number of previous lines of therapy Type of response were reported Best duration of response to a previous single line of therapy Duration since metastasis Year of treatment Number of similar patients treated to observe the response
180/180 (100%) 102/107a (95.3%) 169/180 (93.9%) 165/180 (91.7%) 156/180 (86.7%) 148/180 (822%) 80/133b (60.2%) 71/133b (53.3%) 88/180 (48.9%) 26/180 (14.4%)
Duration since metastasis Z Duration of response from diagnosis of metastatic disease to prior to the start of the exceptional drug. a Of the 180 cases, 107 cases reported at least one driver mutation. b Of the 180 cases, 133 cases used two or more lines of therapy.
of patients (102/180) also had undergone at least one surgical procedure, stem cell transplant or radiation, and 18% of patients (32/180) underwent a procedure directed at treatment of oligometastatic disease. The median best duration of response to a prior therapy was 8 months (range Z 1 to 156) among the 80 cases that reported this. Only 14% (26/180) of cases reported how many others were treated with the same drug to observe this response, and among those the median was 160 patients (range Z 5 to 210). We list the fraction of cases that report pertinent information in Table 1. We note 86% (154/180) of cases did not report the number of patients treated similarly, that is, the number of patients treated with the drug on a protocol or if, off protocol, the number of patients treated with the use of that genetic test (e.g. Foundation Medicine or in-house next generation sequencing (NGS) panel). In other words, some estimate of the denominator from which the case arose. 3.3. Comparing super responses with prior responses For the 106 (of 180) reports that provided duration of response to the exceptional drug as well as past therapies if existed, we constructed a swimmers plot to examine response to therapy since metastatic or unresectable disease (Fig. 1). Of these cases, 32% (34/106) received the exceptional drug as initial therapy. Multiple cases show a substantial duration of response to prior therapies before receipt of the exceptional drug (Fig. 1). When reported, we compared the ratio of the best duration of response to a prior therapy and the duration of response to the present therapy. There were 79 cases that reported this information. Fig. 2 depicts this. Of these cases, 56% (45/79) that showed a PFS to the exceptional drug was greater than a prior PFS. When using the more stringent Von Hoff criteria (which asks for the PFSex/PFSprev 1.3), 49% (39/79) cases demonstrated this characteristic [9].
3.4. Potential financial conflict of interests In 22% (40/180), the authors declared a COI to the drug that was reported to lead to the exceptional response. In 15% (27/180), the authors declared a conflict with the test that was used to characterise an exceptional response. In 57% (n Z 103) of cases, the authors reported a COI to the biomedical industry in general. We noted that case reports with a financial conflict to the test (often a genomic test) were more likely to omit data (missing elements present in no COI [median Z 2, 25th percentile Z 2, 75th percentile Z 3], COI to the test company [median Z 4, 25th percentile Z 2, 75th percentile Z 5], COI to the drug company [median Z 2, 25th percentile Z 1, 75th percentile Z 4] and COI not related to the test or drug company [median Z 2.5, 25th percentile Z 1, 75th percentile Z 3]). Using analysis of variance with a Bonferroni correction for multiple comparisons, we found that articles with a declared COI to the test company had a significantly greater number of missing observations than articles with no declared COI (p-value < .001), COI to the drug company (p-value Z .032) and with COI not related to the test or drug company (p-value Z .019). All other comparisons were not significantly different. 4. Discussion We examined characteristics of case reports of super responders, noting key omissions. Our analysis of the published biomedical literature complements the results of forthcoming prospective studies, such as the NCI’s Exceptional Responders Initiative, and provides guidance regarding a standardised set of reporting criteria of these cases. 4.1. Incomplete reporting To draw conclusions from reports, cases should be well documented. We found that many studies
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
147
Fig. 1. Swimmers plot of cases that reported response since metastatic disease. We identified 106 cases that reported duration of therapy since metastatic or unresectable disease. The graph shows previous total response, if the patient received prior therapy in gray. The colour bars represent response to the super drug per the authors of the article. Symbols represent the final status of the patient. Ambiguous wording of cases were mutually settled on by all the authors collectively. CML, chronic myeloid leukemia; CR, complete response; MDS, myelodysplastic syndrome; PR, partial response, SD, stable disease; NSCLC, nonesmall-cell lung cancer; SCC, squamous cell carcinoma.
148
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
reporting exceptional responders were insufficiently informative and were missing information that could help inform whether the case is actually a super response to a drug. Of all the cases, 51% (92/180) left out the year of treatment. This is important in providing the historical context of when the therapy was started and informs readers what other therapies were standard of care at that time. We noted that duration of response was not reported in 47% (62/133) of cases where patients were treated with multiple lines after metastatic disease, and 40% (53/133) of cases were missing best prior response. These values are useful to report and to compare the exceptional drug with prior drugs to note its relative effectiveness. The
total number of patients treated to observe the exceptional case was reported in a small fraction (14%, 26/180) of cases. This is important in that the denominator can help inform the likelihood an individual would respond to this drug and also shed light on the potential toxicity of this therapy among those who do not respond. Additionally, some cases left out the depth of response (e.g. CR, PR, SD) of the exceptional drug (18%, 32/180), which is critical for evaluating whether a response was present. The lack of complete reporting makes it challenging for clinicians to use this information to guide clinical decision-making and researchers to generate hypotheses for further studying patients who fall into this category.
Fig. 2. A number of cases had a better best prior response compared with response to the exceptional drug. Seventy-nine cases reported both the duration of response to prior lines of therapy and the duration of response to the super drug. This graph represents the PFS of the exceptional drug (PFSex) as compared to the PFS of the best prior response (PFSprev) since metastatic/unresectable disease. The dashed line represents PFSex/PFSprev Z 1. The solid line represents PFSex/PFSprev 1.3. This corresponded to 56% and 49% of the cases, respectively. PFS, progression-free survival.
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
149
Fig. 3. Authors’ financial conflicts of interest. This pie chart notes declared conflicts of interest of any of the authors of the 180 cases we characterised.
4.2. Many cases of exceptional responders had confounding elements
4.3. Exceptional responders are likely a very small set of patients with cancer
For the cases that reported duration of response, including the super response, it was sometimes not clear that the dramatic response was significant in the context of the patient’s disease course. We illustrate this with a swimmer’s plot of all the cases that reported response durations (Fig. 1). The most compelling cases of patients with an exceptional response in Fig. 1 are likely those with a shorter segment for previous total response (in gray) followed by a longer CR (in green), reflecting a true atypical response. On the other hand, some of these cases may represent indolent diseasedi.e. consistently long responses irrespective of the drug. This can be the case even for known aggressive disease states; for example, a Netherlands Cancer Registry retrospective analysis found that the 5-year survival of pathologically verified pancreatic adenocarcinoma was 1.7% [10]. Cases such as those with a prolonged previous total response (in gray) may reflect indolent or pansensitive tumour biology. For example, one notable case we found is a patient with a lung adenocarcinoma who underwent six previous lines of therapy for 65 months before receiving the drugs that led to the exceptional response [11]. Additionally, we noted that of cases that had response durations for at least one other therapy, 44% (34/79) had a longer response to the prior drug compared with the reported super drug. This percentage increased to 51% (40/79) if the more stringent Von Hoff criteria of 1.3 times the prior response were used. Furthermore, we found that a substantial number 32/ 180 (18%) of cases underwent treatment directed at eradicating oligometastatic disease, which can serve to confound exceptional responses seen.
We identified 180 case reports of super response. It is worth putting these roughly in contrast against the number of patients who have undergone next-generation sequencing or comparable commercial testing. At least 120,000 cancer patients have had their tumours profiled through Foundation Medicine, Guardian 360 and the GENIE consortium [12e14]. Although it must be noted, many cases of super response may not be reported in the literature. 4.4. Cases reported few correlative studies to support selection of targeted super drug Of the 274 drugs used as monotherapy or in combination, 161 were targeted therapies. Many of the cases (59%, 107/180) report a driver mutation, with most of these cases reporting the type of genomic analysis used (95%, 102/107). We noted that 43% (46/107) of cases reported just one correlative study. Recently multiple prominent investigations in solid tumours such as lung and renal cell have identified cancers as having significant heterogeneity [15,16]. This makes one question whether simply performing one correlative study will fairly represent a patient’s disease. Additionally, researchers have argued that a prerequisite to understanding the pathophysiology of an individual’s cancer requires -omic, epigenetic and functional studies from multiple sources (tumour, blood, metastatic site) [17,18]. 4.5. Reporting of exceptional responders moving forward We found that 16.1% of our cases (29/180) had duration of response to the exceptional drug of fewer than 6
150
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
Table 2 Recommended characteristics to report in case reports claiming remarkable response to drug. Characteristic
Information that should be reported
Tumor characteristics
Disease Stage
Patient characteristics
Age of patient Last known final status of patient
Characteristics of exceptional drug
Year of treatment Drug used
Cancer type and relevant histology Stage the patient was diagnosed with cancer as well as the stage the patient started the exceptional drug Age the patient was treated with the drug Language clearly stating whether the patient remains alive, developed progressive disease, or died at the time of publication Year the patient started the drug Drug that led to remarkable the response along with all prior lines of therapy and non-drug therapies. Should discuss if this was part of clinical trial. Based on RECIST criteria for solid tumors or disease specific response criteria for hematologic malignancies From time of initiation of drug to disease progression on exceptional drug or status at publication Should include whether the study was conducted as clinical trial or if it is a report of a single exceptional case Should mention how likely the mutation is attributable to the super response, and whether the mutation is new or a known mutation Include test performed, sample tested (blood, primary tumor, metastatectomy etc). Should include test company or where the tests were performed Mention if previous lines include neoadjuvant or adjuvant, or if enrolled in experimental trial Longest response to a previous line of therapy
Type of response
Studies to identify potential driver mutation
Length of documented response to exceptional drug Number of similar patients treated to observe the response Presence of driver mutation
Correlative studies
Prior drug treatment
Other treatment modalities used
Number of previous lines of therapy Best duration of response to a previous single line of therapy Duration of response since metastasis to prior to start of exceptional drug Previous surgery, transplant or radiation Presence of metastectomy or oligometastatic radiotherapy
months. Because the NCI’s Exceptional Responder’s initiative defines that an exceptional responder case must have either a CR or a PR with duration of at least 6 months, those cases would not meet this criterion [19]. Given the importance of complete reporting, we have devised an essential list of items that authors should consider reporting when writing a case report about a patient who has had an exceptional response to a cancer drug (Table 2).This information includes detailed characteristics and history of the patient’s disease course as well as correlative studies. 4.6. Study limitations Our study has several limitations. Although we used several, disparate search strategies, it is possible we missed some articles. Furthermore, not all cases of exceptional responders may be published in the literature. However, we screened over 10,000 publications, which represents the largest effort to date. Second, our reporting criteria were prespecified, but reasonable people may disagree that all of these elements are necessary or alternatively may wish to include aspects that we did not. For this reason, we believe our analysis is merely the starting point for a broader discussion about the core elements that should be reported in all
Should include sites, frequency, and timing of intervention Should include sites, frequency, and timing of intervention
cases of super response. Table 2 is our suggestion for this starting point. 5. Conclusion We found 180 cases of super response to a cancer drug. Many were missing key information about the nature of the response and pertinent past oncologic history. When the current duration of response could be compared with the best prior duration, less than half had a ratio greater than 1.3. To maximise the usefulness of these case reports in the future, standardised reporting criteria are needed. Contributions All authors designed, conducted and drafted the article. Conflict of interest statement Vinay Prasad reports receiving royalties from his book Ending Medical Reversal; that his work is funded by the Laura and John Arnold Foundation; that he has received honoraria for Grand Rounds/lectures from several universities, medical centers, and professional societies and payments for contributions to Medscape.
G. Nishikawa et al. / European Journal of Cancer 101 (2018) 143e151
The other authors declare that they have no conflict of interest to disclose. Authors’ declaration The authors confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. They further confirm that the order of authors listed in the manuscript has been approved by all of them. They confirm that they have given due consideration to the protection of intellectual property associated with this work and that there are no impediments to publication, including the timing of publication, with respect to intellectual property. In so doing they confirm that they have followed the regulations of their institutions concerning intellectual property. They understand that the Corresponding Author is the sole contact for the editorial process (including Editorial Manager and direct communications with the office). He is responsible for communicating with the other authors about progress, submissions of revisions and final approval of proofs.
Acknowledgements Daniel Lecoanet for figure illustration assistance. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.ejca.2018.06.010.
References [1] Hyman DM, Taylor BS, Baselga J. Implementing genome-driven oncology. Cell 2017;168(4):584e99. [2] Takebe N, McShane L, Conley B. Biomarkers: exceptional responders-discovering predictive biomarkers. Nat Rev Clin Oncol 2015;12(3):132e4. [3] Marx V. Cancer: a most exceptional response. Nature 2015; 520(7547):389e93.
151
[4] Printz C. NCI launches exceptional responders initiative: researchers will attempt to identify why some patients respond to treatment so much better than others. Cancer 2015;121(6):803e4. [5] Conley B, Ivy S, Tricoli J, et al. The NCI exceptional responders initiatives: initial feasibility result [Abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research75; 2015 (15 Suppl):Abstract nr 612. [6] Cancer researchers report progress in studying exceptional responders. National cancer Institute NIH; 2017. https://www. cancer.gov/news-events/cancer-currents-blog/2017/exceptionalresponders-progress. [Accessed 26 October 2017]. [7] Iyer G, Hanrahan AJ, Milowsky MI, et al. Genome sequencing identifies a basis for everolimus sensitivity. Science (New York, NY). 2012;338(6104):221. [8] Prasad V, Vandross A. Characteristics of exceptional or super responders to cancer drugs. Mayo Clin Proc 2015;90(12): 1639e49. [9] Von Hoff DD, Stephenson Jr JJ, Rosen P, et al. Pilot study using molecular profiling of patients’ tumors to find potential targets and select treatments for their refractory cancers. J Clin Oncol: Off J Am Soc Clin Oncol 2010;28(33):4877e83. [10] Zijlstra M, Bernards N, de Hingh IH, et al. Does long-term survival exist in pancreatic adenocarcinoma? Acta Oncol (Stockholm, Sweden). 2016;55(3):259e64. [11] Chiari R, Duranti S, Ludovini V, et al. Long-term response to gefitinib and crizotinib in lung adenocarcinoma harboring both epidermal growth factor receptor mutation and EML4-ALK fusion gene. J Clin Oncol: Off J Am Soc Clin Oncol 2014;32(9): e30e2. [12] Foundation Medicine Announces. 2016 second quarter results and recent highlights [press release]. Foundation medicine [Online]. Foundation Medicine, Inc; October 26, 2017 2016. [13] Teichert E. Innovations: startup’s liquid biopsies match cancer patients with trials [Online] Mod Healthc 2016. http://www. modernhealthcare.com/article/20160820/MAGAZINE/ 308209978. [Accessed 26 October 2017]. [14] Consortium APG. AACR project GENIE: powering precision medicine through an international consortium. Canc Discov 2017; 7(8):818e31. [15] Jamal-Hanjani M, Wilson GA, McGranahan N, et al. Tracking the evolution of non-small-cell lung cancer. N Engl J Med 2017; 376(22):2109e21. [16] Gerlinger M, Rowan AJ, Horswell S, et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med 2012;366(10):883e92. [17] Letai A. Functional precision cancer medicine-moving beyond pure genomics. Nat Med 2017;23(9):1028e35. [18] Alizadeh AA, Aranda V, Bardelli A, et al. Toward understanding and exploiting tumor heterogeneity. Nat Med 2015;21(8):846e53. [19] Exceptional responders initiative: questions and answers. National Cancer Institute NIH; 2017. https://www.cancer.gov/aboutcancer/treatment/research/exceptional-responders-initiative-qa. [Accessed 26 October 2017].