Precision Oncology

The term "precision oncology" is used inconsistently and includes the use of targeted therapeutics, new sequencing methods and tissue-agnostic therapy strategies based on molecular alterations. In this guideline, "precision oncology" refers to the use of broad molecular tumor characterization with the aim of personalized therapeutic management.

Molecular testing and molecularly stratified therapies have already become standard for numerous tumor entities (e.g., non-small cell lung carcinoma, colorectal carcinoma, hematological malignancies, etc.). For these cases, please refer to the respective entity-specific guidelines.

However, in addition to guideline-based personalized oncology, effective treatment options can also be identified for patients with other tumor entities through molecular tumor analysis. This is illustrated by the increasing number of predictive biomarkers with cross-entity relevance. Predictive biomarkers can be characterized by alterations in a few defined molecules (BRAF mutations, NTRK fusions) or as complex biomarkers based on the identification of numerous alterations (microsatellite instability (MSI), tumor mutation burden (TMB), or deficient homologous DNA recombination (HRD)). In addition, biomarkers can be detected in various analytes (DNA, RNA, protein) and by an increasing number of (high-throughput) methods. For this reason, broad molecular analyses are often carried out to detect numerous possible biomarkers. In addition, alterations potentially therapeutically relevant but not covered by current guidelines can also be found as part of guideline-based molecular diagnostics.

Adequate indication, test selection and clinical interpretation of molecular alterations with the aim of evidence-based personalized therapy, despite often limited clinical data, is a challenge and requires multidisciplinary expertise within the framework of specialized molecular tumor boards.

The performance and interpretation of broad molecular testing require multidisciplinary expertise. This includes the correct indication, the molecular analysis itself, its informational evaluation, the interpretation of the results and their classification in the clinical context with the aim of treatment recommendations [15]. A schematic sequence of work steps in precision oncology is shown in Figure 1.

If the indication is clear, i.e., based on guidelines and corresponding approval, it is sufficient to present the patient to a multidisciplinary organ tumor board (e.g., typical EGFR mutation in NSCLC, BRAF p.V600E mutation in melanoma, RET mutation in medullary thyroid carcinoma). Complex findings not covered by guidelines, usually in the context of re-biopsies after prior standard treatment, should be presented to a molecular tumor board (MTB).

Outside of clinical trials, the indication for molecular diagnostics should always be based on the possible clinical relevance of the result. The patient's ability to undergo therapy, the current guidelines, the approval status of the therapy that may result from the diagnosis or the latest clinical and scientific findings and, if applicable, the decision of a (molecular) tumor board should be taken into account.

Molecular diagnostics are increasingly being included in guideline-based standard therapy for first-line treatment or relapsed disease and are part of the standard diagnostics for the respective tumor entity. Examples include many driver alterations in NSCLC, BRAF mutations in melanoma and colon carcinoma, HER2 amplifications in breast cancer, RET mutations in medullary thyroid carcinoma and many more. For these patients, following molecular diagnostics, presentation to an organ tumor board is usually sufficient for the treatment decision.

Other indications for molecular diagnostics include the fulfillment of inclusion criteria for molecular-stratified clinical trials. Here, too, molecular diagnostics are increasingly moving into early lines of treatment. For example, the centers participating in the national Network for Genomic Medicine Lung Cancer (nNGM) are integrating study-relevant molecular markers into their NGS-based primary diagnostics in addition to the approved molecular therapy options.

In patients who have failed standard therapies, including approved molecularly targeted therapies, the indication for (possibly extended) molecular diagnostics can be decided by an organ tumor board. The resulting findings should then be discussed in a molecular tumor board. Such molecular tumor boards are established at most comprehensive cancer centers (CCCs) and increasingly also at other cancer centers.

In accordance with the recommendations of the German Network for Personalized Medicine (DNPM), which are also included in the OnkoZert certification criteria for centers for personalized medicine, the following inclusion criteria have been specified for access to molecular tumor boards (with corresponding broad molecular characterization) (Table 1)

The decision to order extended molecular diagnostics after standard therapies have failed usually requires individual decision-making in a multidisciplinary tumor conference. Particularly in patients with rapidly progressive tumor diseases, the choice of the appropriate method, the expected duration of the diagnostic procedures and the subsequent clinical implementation should be taken into account and corresponding analyses should therefore be initiated at an early stage.

Molecular characterization of tumors is increasingly being used as part of routine diagnostics and, based on current guidelines, also at an early stage of treatment, for example before the first-line therapy decision is made. In the majority of these cases, a molecular tumor board is not necessary. However, with the increasing use of methods for parallel DNA and RNA sequencing of numerous gene regions, genetic aberrations whose biological and clinical relevance are not certain (example: atypical EGFR mutations) or complex molecular alterations such as the simultaneous presence of several potentially predictive biomarkers are also found. The clinical classification of such findings, as well as variants of unclear significance (VUS), may require the additional involvement of a molecular tumor board.

Figure 2 illustrates a possible flow chart for integrating broad molecular diagnostics or specialized discussion of findings into the clinical treatment/decision algorithm.

In addition to patient selection (see Table 1), the selection of the appropriate (tumor) material to be examined also plays a role. Both the representativeness of the available sample and the fixation technique must be taken into account. Table 2 lists criteria that can be used when deciding whether to take a new sample.

In addition to re-sampling, liquid biopsy techniques can also be used for some indications. It should be noted that these have a high specificity for many markers, especially fusion genes and amplifications, but (still) have a lower sensitivity compared to tissue biopsies. Tissue diagnostics therefore remain the current standard, particularly in primary diagnostics, while liquid biopsy is increasingly being used in recurrence settings where biopsies are difficult to obtain and is also increasingly being used for therapy monitoring as part of clinical studies.

Therapeutically relevant molecular alterations can be identified at numerous levels of cellular regulation. Table 3 summarizes possible levels of cellular dysregulation.

New technologies enable the rapid and simultaneous analysis of numerous possible biomarkers using high-throughput methods. In the field of precision oncology, DNA and RNA sequencing currently play the most important role alongside immunohistochemistry. Possible diagnostic procedures are summarized in Table 4 [1, 8].

The selection of a suitable molecular diagnostic test should also take into account previous analyses and the pre-test probability of relevant biomarkers (Tables 8 and 9).

The performance of molecular testing and identification of molecular alterations is a largely standardized process and the domain of human genetics, (molecular) pathology and bioinformatics, among others, and should be carried out in a quality-assured environment, if possible, in an institute/laboratory accredited for molecular diagnostics. Ideally, molecular diagnostics should be embedded in a clinical environment that ensures the necessary multidisciplinarity with regard to the interpretation and clinical annotation of the results.

The report should include information on the (tumor) material used, material quality, tumor cell content, type of analysis performed, listing of identified alterations in standardized coding, variant allele frequency and a functional evaluation of the variants. Human genetics must be included in the evaluation of potential germline variants.

The functional assessment of molecular alterations should be carried out on the basis of existing guidelines and SOPs (e.g., [6]).

Specific molecular alterations may contain relevant informations for the clinical management of malignant diseases. The identification of these biomarkers is therefore the goal of molecular testing. The clinical annotation of a variant serves to evaluate it as a biomarker. The following possible biomarkers can be distinguished (Table 5) [16].

An annotation as a predictive biomarker often requires an extensive literature search. Numerous databases allow a simplified search for information. The content of the databases often does not overlap (e.g. civicdb.org, oncokb.org, https://ckb.jax.org/) and analysis of the primary literature is essential to classify the findings [14].

To evaluate the corresponding biomarker against the background of the respective tumor entity, evidence levels were defined, which should be used for assessment [7, 10, 13]. In German-speaking countries, the NCT/ZPM evidence levels are the most widely used. In addition to these, there are other evidence levels such as AMP/ASCO/ACP [11] or ESMO-ESCAT [12].

In addition to the interpretation of potential predictive biomarkers, other relevant biomarkers should also be assessed (e.g., DPYD diagnostics, if included). In the case of germline diagnostics, this requires the expertise and co-assessment of human genetics. Due to the high time expenditure and the necessary research work, the clinical classification and interpretation of the findings often takes place before the final decision on clinical application.

There are an increasing number of predictive biomarkers with tissue-agnostic approval for targeted drugs (corresponding to at least ESCAT Level I-C, NCT m1A, JCR Tier 1 A.1). Corresponding predictive biomarkers with FDA approval are shown in Table 7. In Europe, there are currently two tissue-agnostic approvals for NTRK inhibitors and RET inhibitors (NTRK1-3 fusions or RET fusions).

Table 8 summarizes the prevalence of these biomarkers (see Table 7) in various tumor entities.

The final application of the identified and assessed biomarkers in the specific clinical situation requires broad multidisciplinary expertise and should be carried out by designated or certified molecular tumor boards. According to the criteria of the German Network for Personalized Medicine (DNPM; [9]), the participating disciplines of an MTB team include at least hematology and medical oncology, pathology, molecular pathology, molecular biology, bioinformatics, human genetics as well as case-dependent specialist disciplines and radiology as required. In addition, this team should have sufficient and comprehensive expertise in the interpretation, evaluation and classification of molecular findings in the clinical course of therapy, which is ensured by a sufficient number of MTB patients.

The high degree of individual variation and the different levels of evidence for the resulting recommendations require close attention to the clinical course and careful consideration when integrating any experimental treatment options into further therapy.

Ideally, non-approved treatment options should be administered within the framework of prospective clinical studies; the availability of such studies should therefore be evaluated as part of the treatment recommendation and assessed at least on a national level. However, implementation of the treatment recommendation often requires off-label use of medication. Here, the level of evidence for the recommended therapy should be taken into particular consideration with regard to the expected patient benefit. Off-label therapies should be followed up as part of registry studies (see below).

The high frequency of non-approved therapy recommendations and lack of clinical data supports the integration of research-based patient care in precision oncology. In particular, implementation rates of targeted therapies and treatment response to off-label therapies should be extensively tracked and documented in a structured manner (see core data sets of the Medical Informatics Initiative). Ideally, this should be done as part of a prospective registry study and data should be pooled in a data network, as several scientific networks are currently making important contributions here. Consistent recording of molecular alterations can therefore also lead to an improved understanding of molecular pathomechanisms and effective treatment options also for less common malignancies in the future.

1. Avila M, Meric-Bernstam F. Next-generation sequencing for the general cancer patient. Clin Adv Hematol Oncol 2019;17:447-454. PMID:31449513

2. Bonneville R, Krook MA, Kautto EA et al. Landscape of microsatellite instability across 39 cancer types. JCO Precis Oncol 2017;2017:PO.17.00073. DOI:10.1200/PO.17.00073

3. Cancer Genome Atlas Network. Genomic classification of cutaneous melanoma. Cell 2015;161:1681-1696. DOI:10.1016/j.cell.2015.05.044

4. Chakravarty D, Johnson A, Sklar J et al. Somatic genomic testing in patients with metastatic or advanced cancer: ASCO provisional clinical opinion. J Clin Oncol 2022;40:1231-1258. DOI:10.1200/JCO.21.02767

5. Gouda MA, Nelson BE, Buschhorn L, Wahida A, Subbiah V. Tumor-agnostic precision medicine from the AACR GENIE database: clinical implications. Clin Cancer Res 2023;29:2753-2760. DOI:10.1158/1078-0432.CCR-23-0090

6. Horak P, Griffith M, Danos AM et al. Standards for the classification of pathogenicity of somatic variants in cancer (oncogenicity): Joint recommendations of Clinical Genome Resource (ClinGen), Cancer Genomics Consortium (CGC), and Variant Interpretation for Cancer Consortium (VICC). Genet Med 2022;24:986-998. DOI:10.1016/j.gim.2022.01.001

7. Horak P, Leichsenring J, Goldschmid H et al. Assigning evidence to actionability: An introduction to variant interpretation in precision cancer medicine. Genes Chromosomes Cancer 2022;61:303-313. DOI:10.1002/gcc.22987

8. Hussen BM, Abdullah ST, Salihi A et al. The emerging roles of NGS in clinical oncology and personalized medicine. Pathol Res Pract 2022;230:153760. DOI:10.1016/j.prp.2022.153760

9. Illert AL, Stenzinger A, Bitzer M et al. The German Network for Personalized Medicine to enhance patient care and translational research. Nat Med 2023;29:1298-1301. DOI:10.1038/s41591-023-02354-z

10. Leichsenring J, Horak P, Kreutzfeldt S et al. Variant classification in precision oncology. Int J Cancer 2019;145:2996-3010. DOI:10.1002/ijc.32358

11. Li MM, Datto M, Duncavage EJ et al. Standards and guidelines for the interpretation and reporting of sequence variants in cancer: a joint consensus recommendation of the Association for Molecular Pathology, American Society of Clinical Oncology, and College of American Pathologists. J Mol Diagn 2017;19:4-23. DOI:10.1016/j.jmoldx.2016.10.002

12. Mateo J, Chakravarty D, Dienstmann R et al. A framework to rank genomic alterations as targets for cancer precision medicine: the ESMO Scale for Clinical Actionability of molecular Targets (ESCAT). Ann Oncol 2018;29:1895-1902. DOI:10.1093/annonc/mdy263

13. Mock A, Teleanu MV, Kreutzfeldt S et al. NCT/DKFZ MASTER handbook of interpreting whole-genome, transcriptome, and methylome data for precision oncology. NPJ Precis Oncol 2023;7:109. DOI:10.1038/s41698-023-00458-w

14. Pallarz S, Benary M, Lamping M et al. Comparative analysis of public knowledge bases for precision oncology. JCO Precis Oncol 2019;3:PO.18.00371. DOI:10.1200/PO.18.00371

15. Rieke DT, de Bortoli T, Horak P et al. Feasibility and outcome of reproducible clinical interpretation of high-dimensional molecular data: a comparison of two molecular tumor boards. BMC Med 2022;20:367. DOI:10.1186/s12916-022-02560-5

16. Tortora G, Sessa C, Scarpa A, Banerjee S (eds.) ESMO Handbook of translational research, 2^nd edition. ESMO Press Viganello-Lugano 2015.

17. Yan M, Schwaederle M, Arguello D, Millis SZ, Gatalica Z, Kurzrock R. HER2 expression status in diverse cancers: review of results from 37,992 patients. Cancer Metastasis Rev 2015 Mar;34:157-164. DOI:10.1007/s10555-015-9552-6