Despite rare cancers accounting for 25% of adult tumors1, they are difficult to study due to the low disease incidence and geographically dispersed patient populations, which has resulted in significant unmet clinical needs for patients with rare cancers. We assessed whether a patient-partnered research approach using online engagement can overcome these challenges, focusing on angiosarcoma, a sarcoma with an annual incidence of 300 cases in the United States. Here we describe the development of the Angiosarcoma Project (ASCproject), an initiative enabling US and Canadian patients to remotely share their clinical information and biospecimens for research. The project generates and publicly releases clinically annotated genomic data on tumor and germline specimens on an ongoing basis. Over 18 months, 338 patients registered for the ASCproject, which comprises a large proportion of all patients with angiosarcoma. Whole-exome sequencing (WES) of 47 tumors revealed recurrently mutated genes that included KDR, TP53, and PIK3CA. PIK3CA-activating mutations were observed predominantly in primary breast angiosarcoma, which suggested a therapeutic rationale. Angiosarcoma of the head, neck, face and scalp (HNFS) was associated with a high tumor mutation burden (TMB) and a dominant ultraviolet damage mutational signature, which suggested that for the subset of patients with angiosarcoma of HNFS, ultraviolet damage may be a causative factor and that immune checkpoint inhibition may be beneficial. Medical record review revealed that two patients with HNFS angiosarcoma had received off-label therapeutic use of antibody to the programmed death-1 protein (anti-PD-1) and had experienced exceptional responses, which highlights immune checkpoint inhibition as a therapeutic avenue for HNFS angiosarcoma. This patient-partnered approach has catalyzed an opportunity to discover the etiology and potential therapies for patients with angiosarcoma. Collectively, this proof-of-concept study demonstrates that empowering patients to directly participate in research can overcome barriers in rare diseases and can enable discoveries.
Subscribe to Journal
Get full journal access for 1 year
only $18.75 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
To protect patient confidentiality, the study data set is de-identified before it is shared, which includes the masking of patient IDs and the reclassification of unique patient-reported demographic responses as ‘other’ (see Supplementary Methods). The resulting clinically annotated genomic data set of the ASCproject is shared publicly on cBioPortal on an ongoing and regular basis as the data are generated; the ASCproject has been registered as a study at dbGaP under the accession number phs001931 (contact email@example.com with any questions regarding data availability).
All software and pipelines for genomic data generation are described in detail in Supplementary Methods. Information on the scripts used to generate the figures is accessible upon request from the corresponding author.
Sharifnia, T., Hong, A. L., Painter, C. A. & Boehm, J. S. Emerging opportunities for target discovery in rare cancers. Cell Chem. Biol. 24, 1075–1091 (2017).
Lahat, G. et al. Angiosarcoma: clinical and molecular insights. Ann. Surg. 251, 1098–1106 (2010).
Hoang, N. T., Acevedo, L. A., Mann, M. J. & Tolani, B. A review of soft-tissue sarcomas: translation of biological advances into treatment measures. Cancer Manag. Res. 10, 1089–1114 (2018).
Cohen-Hallaleh, R. B. et al. Radiation induced angiosarcoma of the breast: outcomes from a retrospective case series. Clin. Sarcoma Res. 7, 15 (2017).
Lee, F. I. & Harry, D. S. Angiosarcoma of the liver in a vinyl-chloride worker. Lancet 1, 1316–1318 (1974).
Behjati, S. et al. Recurrent PTPRB and PLCG1 mutations in angiosarcoma. Nat. Genet. 46, 376–379 (2014).
Guo, T. et al. Consistent MYC and FLT4 gene amplification in radiation-induced angiosarcoma but not in other radiation-associated atypical vascular lesions. Genes Chromosomes Cancer 50, 25–33 (2011).
Arora, T. K., Terracina, K. P., Soong, J., Idowu, M. O. & Takabe, K. Primary and secondary angiosarcoma of the breast. Gland Surg. 3, 28–34 (2014).
Giacomini, C. P. et al. Breakpoint analysis of transcriptional and genomic profiles uncovers novel gene fusions spanning multiple human cancer types. PLoS Genet. 9, e1003464 (2013).
Murali, R. et al. Targeted massively parallel sequencing of angiosarcomas reveals frequent activation of the mitogen activated protein kinase pathway. Oncotarget 6, 36041–36052 (2015).
Kunze, K. et al. A recurrent activating PLCG1 mutation in cardiac angiosarcomas increases apoptosis resistance and invasiveness of endothelial cells. Cancer Res. 74, 6173–6183 (2014).
Shon, W., Sukov, W. R., Jenkins, S. M. & Folpe, A. L. MYC amplification and overexpression in primary cutaneous angiosarcoma: a fluorescence in-situ hybridization and immunohistochemical study. Mod. Pathol. 27, 509–515 (2014).
Lawrence, M. S. et al. Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature 499, 214–218 (2013).
Italiano, A. et al. Alterations of the p53 and PIK3CA/AKT/mTOR pathways in angiosarcomas: a pattern distinct from other sarcomas with complex genomics. Cancer 118, 5878–5887 (2012).
Chang, M. T. et al. Accelerating discovery of functional mutant alleles in cancer. Cancer Discov. 8, 174–183 (2018).
Hon, W. C., Berndt, A. & Williams, R. L. Regulation of lipid binding underlies the activation mechanism of class IA PI3-kinases. Oncogene 31, 3655–3666 (2012).
Dogruluk, T. et al. Identification of variant-specific functions of PIK3CA by rapid phenotyping of rare mutations. Cancer Res. 75, 5341–5354 (2015).
Gymnopoulos, M., Elsliger, M. A. & Vogt, P. K. Rare cancer-specific mutations in PIK3CA show gain of function. Proc. Natl Acad. Sci. USA 104, 5569–5574 (2007).
Ikenoue, T. et al. Functional analysis of PIK3CA gene mutations in human colorectal cancer. Cancer Res. 65, 4562–4567 (2005).
Andre, F. et al. Alpelisib for PIK3CA-Mutated, hormone receptor-positive advanced breast cancer. N. Engl. J. Med. 380, 1929–1940 (2019).
Juric, D. et al. Alpelisib plus fulvestrant in PIK3CA-altered and PIK3CA-wild-type estrogen receptor-positive advanced breast cancer: a phase 1b clinical trial. JAMA Oncol. 5, e184475 (2018).
Juric, D. et al. Phosphatidylinositol 3-kinase alpha-selective inhibition with alpelisib (BYL719) in PIK3CA-altered solid tumors: results from the first-in-human study. J. Clin. Oncol. 36, 1291–1299 (2018).
Kandoth, C. et al. Mutational landscape and significance across 12 major cancer types. Nature 502, 333–339 (2013).
Ciriello, G. et al. Comprehensive molecular portraits of invasive lobular breast cancer. Cell 163, 506–519 (2015).
Alexandrov, L. B. et al. Signatures of mutational processes in human cancer. Nature 500, 415–421 (2013).
Cancer Genome Atlas Research Network. Comprehensive and integrated genomic characterization of adult soft tissue sarcomas. Cell 171, 950–965 e28 (2017).
Campbell, B. B. et al. Comprehensive analysis of hypermutation in human cancer. Cell 171, 1042–1056 e10 (2017).
Zehir, A. et al. Mutational landscape of metastatic cancer revealed from prospective clinical sequencing of 10,000 patients. Nat. Med. 23, 703–713 (2017).
Griewank, K. G. et al. TERT promoter mutations are frequent in atypical fibroxanthomas and pleomorphic dermal sarcomas. Mod. Pathol. 27, 502–508 (2014).
Mullins, B. & Hackman, T. Angiosarcoma of the head and neck. Int. Arch. Otorhinolaryngol. 19, 191–195 (2015).
Karube, R. et al. Angiosarcoma of the scalp diagnosed by the presence of neck inflammation: a case report. Int. J. Oral Sci. 4, 166–169 (2012).
Samstein, R. M. et al. Tumor mutational load predicts survival after immunotherapy across multiple cancer types. Nat. Genet. 51, 202–206 (2019).
Conway, J. R., Kofman, E., Mo, S. S., Elmarakeby, H. & Van Allen, E. Genomics of response to immune checkpoint therapies for cancer: implications for precision medicine. Genome Med. 10, 93 (2018).
Rizvi, N. A. et al. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer. Science 348, 124–128 (2015).
Hodis, E. et al. A landscape of driver mutations in melanoma. Cell 150, 251–263 (2012).
Snyder, A. et al. Genetic basis for clinical response to CTLA-4 blockade in melanoma. N. Engl. J. Med. 371, 2189–2199 (2014).
Van Allen, E. M. et al. Genomic correlates of response to CTLA-4 blockade in metastatic melanoma. Science 350, 207–211 (2015).
van Rooij, N. et al. Tumor exome analysis reveals neoantigen-specific T-cell reactivity in an ipilimumab-responsive melanoma. J. Clin. Oncol. 31, e439–e442 (2013).
Levine, R. L. et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis. Cancer Cell 7, 387–397 (2005).
Girard, N. et al. Analysis of genetic variants in never-smokers with lung cancer facilitated by an internet-based blood collection protocol: a preliminary report. Clin. Cancer Res. 16, 755–763 (2010).
We thank the many patients with angiosarcoma and loved ones of patients who have generously partnered with us to create and drive this research project; we are grateful to work with you every day. We thank the ASCproject advocacy partners (Angiosarcoma Awareness, The Paula Takacs Foundation for Sarcoma Research, Sarcoma Alliance for Research through Collaboration, Sarcoma Alliance, Sarcoma Foundation of America, The Sarcoma Coalition, and Target Cancer Foundation). We thank K. Shanahan for her assistance with medical record abstraction. We thank colleagues from across the Broad Institute and Dana Farber for helpful scientific discussions and support. We thank W. Hahn for helpful feedback on the manuscript. We thank the Broad Institute Communications & Development teams for their hard work to support this project. We are especially thankful to all members of the Count Me In team, the Wagle laboratory, the engineering team at the Broad Institute (A. Zimmer, E. Baker, S. Maiwald, J. Lapan, S. Sutherland), the Broad Institute Cancer Program, the Broad Institute Genomics Platform, and the compliance team at the Broad Institute. This research was supported by anonymous philanthropic support to the Broad Institute.
C.A.P. is a nominal stockholder in Supernus Pharmaceuticals. C.A.P. has received sponsored research support from Eisai Inc. J.L.H. is a consultant to Eli Lilly and Epizyme. G.D.D. reports the following interests: grants, personal fees, non-financial support and travel support to consulting meetings from Novartis, Bayer, Roche, Epizyme and Daiichi-Sankyo; grants, personal fees and travel support to consulting meetings from Pfizer; personal fees and travel support to consulting meetings from EMD-Serono; personal fees from Sanofi; grants and personal fees from Ignyta; grants, personal fees and travel support to consulting meetings from Loxo Oncology; grants, personal fees and non-financial support from AbbVie; personal fees and travel support to consulting meetings from Mirati Therapeutics; personal fees and travel support to consulting meetings from WIRB Copernicus Group; personal fees from ZioPharm; personal fees from Polaris Pharmaceuticals; personal fees and travel support to consulting meetings from M.J. Hennessey/OncLive; grants, personal fees and travel support to consulting meetings from Adaptimmune; grants from GlaxoSmithKline; personal fees, minor equity, and travel support to Board meetings from Blueprint Medicines, where he serves as a member of the Board of Directors; personal fees and minor equity options from Merrimack Pharmaceuticals, where he serves as a member of the Board of Directors; personal fees and minor equity from G1 Therapeutics; personal fees, minor equity options and travel support to consulting meetings from CARIS Life Sciences; minor equity options from Bessor Pharmaceuticals; minor equity options from ERASCA Pharmaceuticals; personal fees and travel support to consulting meetings from CHAMPIONS Oncology; grants and personal fees from Janssen; grants, personal fees, travel support to consulting meetings, and non-financial support from PharmaMar. In addition, G.D.D. has a use patent on imatinib for GIST, licensed to Novartis with royalties paid to the Dana-Farber Cancer Institute. E.S.L. serves on the Board of Directors for Codiak BioSciences and Neon Therapeutics, and serves on the Scientific Advisory Board of F-Prime Capital Partners and Third Rock Ventures; he is also affiliated with several non-profit organizations including serving on the Board of Directors of the Innocence Project, Count Me In, and Biden Cancer Initiative, and the Board of Trustees for the Parker Institute for Cancer Immunotherapy. E.S.L. has served and continues to serve on various federal advisory committees. T.R.G. serves or has recently served as a scientific adviser to Foundation Medicine, Inc. (wholly owned by Roche), GlaxoSmithKline, plc, Sherlock Biosciences, Inc., and FORMA Therapeutics, Inc. N.W. was previously a stockholder and consultant for Foundation Medicine, Inc., has been a consultant/advisor for Novartis and Eli Lilly, and has received sponsored research support from Novartis and Puma Biotechnology. None of the for-profit entities had any role in the conceptualization, design, data collection, analysis, decision to publish, or preparation of the manuscript.
Peer review information Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
(a) Patient intake survey, before (top) and after (bottom) incorporation of patient feedback. (b) A summary of a social media post by study staff soliciting feedback on potential home page designs and comments from the community. (c) The turtle mascot associated with angiosarcoma which, based on patient feedback, was custom designed by the study and is prominently displayed on ASCproject.org as well as on saliva and blood kits. (d) A summary of a social media post by study staff polling angiosarcoma patients for interest in participating in ASCproject.org prior to its development. Ninety patients responded affirmatively within days.
(a) A summary of a social media post by a patient raising awareness for ASCproject.org and encouraging new patients to visit the project website. (b) A list of advocacy partners supporting ASCproject.org. (c) A summary of a social media post by an advocacy partner raising awareness for the project through their existing network. (d) A summary of a social media post by a patient sharing the value that participation in research holds for them personally, and encouraging others to learn more about the ASCproject. (e) A summary of a social media post by an advocacy partner supporting ASCproject.org. (f) A summary of a social media post by an advocacy partner promoting a scientific poster presentation and raising awareness for ASCproject.org.
(a) A summary of one of 110 social media posts by study staff providing study progress updates. A map was shared representing the first 110 patients to join the study as well as people who completed a survey about a loved one who passed away from angiosarcoma. (b) A summary of a live-stream social media post by study staff conducting a lay-friendly video walkthrough of a poster presented at a scientific conference, and comments from angiosarcoma community members. The copyright of the image is owned by a member of the study staff and is shared here with permission. (c) A summary of a social media post by study staff sharing an update on an upcoming opportunity to present about the ASCproject at an NIH meeting, and comments from angiosarcoma community members. (d) A summary of a social media post by study staff sharing the aggregate results of patient-reported information provided by patients diagnosed with radiation-induced angiosarcoma of the breast in response to questions about the sub-group from the community.
Extended Data Fig. 4 Additional patient-reported data from the 227 consented AS patients in the Angiosarcoma Project.
(a) This chart depicts the 340 different clinical institutions that ASCproject patients reported they received some aspect of their clinical care for AS (x-axis) and the number of patients that reported care at any given institution (y-axis). This information was provided on the medical release forms completed by consented Angiosarcoma Project patients (n = 225 patients). Of the 227 consented ASCproject participants, 2 people did not fill out the medical release form, and are not included in this figure. The majority of the 340 total institutions were listed by only one patient. Ten institutions were listed by five or more AS patients as sites of care (box, upper right). (b) A bar chart showing the patient responses to the intake survey question ‘Please select all of the places in your body that you have ever had angiosarcoma (Please select all that apply).’ Patient intake surveys completed by the 227 patients who consented for the Angiosarcoma Project as of September 30, 2018 were analyzed. Each response of a location is counted separately, such that multiple locations are shown for patients that selected more than one answer to indicate that they have had more than one location for their angiosarcoma lesions. 9 patients selected the provided response option of ‘I don’t know’ (‘Don’t Know’), and 7 patients did not respond to this question (’Not Reported’). The HNFS category depicted above includes patient responses of ‘head, face, neck’ and responses of ‘scalp.’ (c) Patient-reported information regarding other cancer diagnoses prior to angiosarcoma. Patient intake surveys completed by the 227 patients who consented for the Angiosarcoma Project as of September 30, 2018 were analyzed to obtain patient-reported information regarding other cancer diagnoses prior to angiosarcoma. Seventy five patients’ responses indicated they had been diagnosed with another cancer prior to angiosarcoma. These patients were identified based on their responses of “Yes” to the survey question “Were you ever diagnosed with any other kind of cancer(s)?”. Patients also provided specific diagnoses years for AS and other cancer(s), which were used to determine that the AS diagnosis occurred in the same year or after another cancer diagnosis. Of those 75 patients, 55 also responded “Yes” to the survey question “Have you had radiation as a treatment for another cancer(s)?”. Of those 55 patients, 39 also reported “breast” as their only primary AS site and indicated breast cancer as a previous cancer. The majority of these 39 patients are expected to be cases of cutaneous AS of the breast. AS, angiosarcoma; ASCproject, The Angiosarcoma Project; HNFS, head, neck, face, scalp.
This diagram shows additional information regarding the attrition at various steps in the Angiosarcoma Project. Numbers indicated are as of September 30, 2018. 59 patients who signed the consent form did not provide a country of residence or indicated they were living outside of the U.S. or Canada. 2 patients did not sign the release form. 2 patients were not sent blood or saliva kits, including one patient who passed away before kits could be sent and another who provided an invalid mailing address. 43 patients did not return either their blood or saliva kits. 39 of 225 patients who signed the medical release form were in the study staff’s medical record request queue as of September 30, 2018. 200 of 419 requested medical records were not received (55 requests resulted in denials by medical record departments, and 145 requests did not get a response). 61 patients’ records either did not contain sufficient information to request tissue or showed too little tissue to request for research. 21 requested FFPE samples have not yet been received. 5 received FFPE samples did not have available matched normal DNA (from blood or saliva) and were not initiated for sequencing. 28 submitted paired samples have not been sequenced (16 samples had insufficient material for sequencing and 12 sets of samples are still in the sequencing pipeline). 21 sequenced samples had less than 10% tumor purity. 2 samples were excluded from the dataset because centralized pathology review determined the samples were not angiosarcoma. The remaining 47 FFPE samples from 36 patients that underwent whole exome sequencing comprised the Angiosarcoma Project September 2018 dataset that was released on cBioPortal.org along with associated patient-reported and clinical data.
Extended Data Fig. 6 Patient-reported data from the 36 AS patients whose samples were sequenced in the Angiosarcoma Project.
The intake survey completed during the ASCproject registration process by these 36 patients were analyzed. (a) A bar chart showing the age in years of these 36 patients at initial diagnosis with AS (mean: 47.8 years). These values were calculated from patient provided date of birth and date of initial AS diagnosis. (b) A bar chart showing the years elapsed between these 36 patients’ initial diagnosis with AS and patients’ registration in the ASCproject (mean: 4.6 years). These values were calculated from the date of project registration and the patient-provided date of initial AS diagnosis. (c) A bar chart showing the patient-reported location of angiosarcoma at the time of last intake survey completion. An option was provided for patients to report no evidence of disease. Patients with more than one location of AS were able to provide more than one site. 2 patients did not respond to this question (‘Not Reported’) and 1 patient responded ‘Don’t Know.’ AS, Angiosarcoma; ASCproject, The Angiosarcoma Project.
(a) Co-mutation plot shows significantly recurring mutated genes among (N=36 patients) AS patients. TP53 and KDR are significantly mutated across the cohort. The p-values were computed using fisher’s method and truncated product method. FDR (q values) were generated using Benjamini and Hochber method to correct for multiple hypotheses. Genes that have the –log 10 q value >=1 (red line) are significant. Box plot of the allelic fractions of the mutations in individual tumors is shown below the mutation plot. The horizontal red line represents median and the whiskers to extend to the data extremes. (b) Stick plot of KDR showing the recurrent mutations and their positions that were identified in this AS cohort. (c) Bar graph representing the number of tumor samples (y-axis) that are grouped based on 5 dominant mutational signature categories (x-axis). (d) Box plot representing the distribution of TMB (y-axis) for N=47 tumor samples across five mutational signature processes (as defined by COSMIC) identified in this cohort and categorized among 8 AS subclassifications (x-axis). Samples with dominant signature 7 (yellow, N=10 tumor samples) which corresponds to UV light exposure, have the highest median TMB and this signature was only observed in HNFS AS. Horizontal bars indicate median values, while the boxes show the IQR. The whiskers extend to 1.5× the IQR on either side. AS, Angiosarcoma; TMB, tumor mutation burden, HNFS; head, neck, face, scalp.
Sensitivity to CRISPR knockout-induced loss of PIK3CA (PIK3CA dependency) was calculated for three groups of cancer cell lines using the Dependency Map dataset (depmap.org): lines containing (1) wild-type PIK3CA (gray), (2) PIK3CA hotspot mutations (red), and (3) PIK3CA mutations seen in AS patients (R88Q [4 lines], P124L [1 line], G914R [1 line]; (colored)). PIK3CA hotspot mutations were defined (depmap.org) using their frequency of occurrence in TCGA and COSMIC databases. Relative to cell lines with wild-type PIK3CA, PIK3CA-mutant cell lines are significantly more sensitive to CRISPR knockout-induced loss of PIK3CA (PIK3CA hotspot mutant (n=64 cell lines, median=-0.669, minimum=-1.16, maximum=-0.147) vs PIK3CA WT (n=479 cell lines, median=-0.298, minimum=-0.961, maximum=0.135), p-value = 1.46 x 10-14, mean difference = -0.328, 95% CI: (-0.396,-0.260)); cell lines with PIK3CA with AS mutations (n=6 cell lines, median=-0.511, minimum=-0.740, maximum=-0.344) vs PIK3CA WT (n=479 cell lines, median=-0.298, minimum=-0.961, maximum=0.135), (p-value = 4.40 x 10-2, mean difference = -0.181, 95% CI: (-0.355,-0.007)). Statistical comparisons were performed using a two-sample, two-sided unpaired t-test. WT, wild-type; AS, angiosarcoma.
A pie chart showing information regarding radiation treatments received by the 36 sequenced patients in the cohort. This information was abstracted from obtained medical records. These radiation therapies may have been given in conjunction with pharmacological and/or surgical interventions. Patients were categorized into groups depicted in pie chart: patients who received neoadjuvant radiation (2 patients), patients who received adjuvant radiation (12 patients), patients who received radiation for AS without surgery (2), and patients who did not receive radiation for AS (19). For one patient who was categorized as ‘Unknown’, the medical records were not sufficient to abstract this information. AS, angiosarcoma.
A pie chart showing patient responses to the patient intake survey question ‘How did you hear about The Angiosarcoma Project?.’ Patient intake surveys completed by the 227 patients who consented for the Angiosarcoma Project as of September 30, 2018 were analyzed. Each free text patient response was grouped into the categories depicted in pie chart. Unique responses from 2 patients were grouped together as ‘Other.’ 4 patients reported more than one referral source (’Multiple’). 6 patients did not respond to the question (’None Listed’).
About this article
Cite this article
Painter, C.A., Jain, E., Tomson, B.N. et al. The Angiosarcoma Project: enabling genomic and clinical discoveries in a rare cancer through patient-partnered research. Nat Med 26, 181–187 (2020). https://doi.org/10.1038/s41591-019-0749-z