In 2000, more than two decades ago, genome-wide gene expression profiling became available and thereafter led to the dissection of cancer biology across almost all entities [1,2,3]. First, the molecular portraits based on RNA expression profiling (termed “heat maps”) were used in breast cancer to identify luminal, ERBB2-positive, and basal tumors. Interestingly, these subtypes not only elucidated the underlying biology but also directly suggested targeted treatment intervention with luminal tumors being hormone-dependent, ERBB2-positive tumors exposing the transmembrane receptor Her-2/neu and basal tumors lacking homogenous expression of typical targeted treatment options, with the latter being termed “triple negative” later on. Interestingly, genome-wide mutation analysis later on revealed that the luminal subtype, while bearing most mutations (such as PIK3CA) exhibited lowest immunogenicity and frequently absence of tumor-infiltrating lymphocytes. In contrast, the basal subtype turned out to have lowest rate of classical oncogens, but was dominated by loss-of-function mutation of p53 [4], while almost half of basal tumors being infiltrated by large amounts of immune cells. This led to the assumption that hormone regulation affects immune cell recognition and three biological axes (hormone, immune, and proliferation axis) were built up for breast cancer as being the coordinates of the biological universe of breast cancer [5, 6]. The therapeutic implication of these fundamental insights were further explored and validated the distinct sensitivity towards antihormonal treatment, ERBB2 targeting, and chemotherapy. Interestingly, the hormone-insensitive, highly proliferating basal and ERBB2-positive tumors with higher amounts of immune cell infiltrates did respond best to neoadjuvant treatment with superior outcome [7]. As one consequence, the concept arose to develop RNA-based vaccination concepts in the post-neoadjuvant situation of triple negative breast cancer not responding to neoadjuvant chemotherapy by targeting individual neo-epitope patterns [8], which has been investigated in the subsequent “Merit” trial with positive proof of concept [9]. In line with this, the first approval of checkpoint therapy treatment in breast cancer happened in the triple negative breast cancer subtype [10].

Almost 10 years after their first description in 2000, the molecular subtypes of breast cancer became integral part for patient stratification in breast cancer by semiquantitative recapitulation using conventional immune histochemistry methods [11] or by molecular methods using standard PCR methods to quantify key targets after RNA extraction from routinely fixed tissues using the in vitro diagnostic “MammaTyper®” test system [12,13,14].

As a next step, this new IVD technology was validated in other disease entities in which molecular subtyping initially identified in breast cancer just started to be recognized as being potentially hormone-driven such as ovarian cancer [15, 16], lung cancer, [17, 18] and bladder cancer [19,20,21].

Importantly the quantitative determination or the main drug targets in breast cancer, that is, estrogen receptor (= ESR1, gene name) and the receptor-tyrosine kinase HER-2/neu (= ERBB2; gene name) revealed that only high mRNA overexpression of the targets is associated with addiction to the target and respective response and efficacy to the treatment. As one example in the NSABP B14 breast cancer trial comparing 5-year tamoxifen vs. placebo in ER-positive tumors by IHC, only tumors with high ESR1 mRNA expression did benefit from the antihormonal treatment, while immunohistochemical staining failed to be predictive [22]. Moreover, the large NSABP P1 prevention trial validated that the benefit of Tamoxifen treatment was restricted to the prevention of very high ESR1 mRNA expression [21]. Similarly, for ERBB2 targeting by the two antibodies Tratuzumab and Pertuzumab within the neoadjuvant TRYPHAENA trial, a large translational program revealed that ERBB2 overexpression remained to be the only marker for patient selection of anti-ERBB2 treatments and therapy benefit prediction [23]. Apparently molecular in vitro diagnostics in breast cancer teaches us that it is the quantitation of the treatment target which is of utmost importance for therapy guidance and precision of treatment efficacy prediction.

Moreover, this directly leads to one of the hallmarks of in vivo diagnostics/theranostics, which presumes that uptake of radioactive ligands is strongly correlated to receptor density on the surface tissue. We therefore evaluated whether the surface expression of SSTR2 receptors as determined by semiquantitative IHC and fully quantitative PCR methods in vitro might be related to the uptake of SSTR2 ligands (DOTA-TOC, DOTA-NOC and DOTA-TATE) in patients suffering neuroendocrine pancreatic tumors [24]. It turned out that conventional IHC methods by immune reactivity score (IRS) only trended to predict uptake as determined by positive correlation with SUV mean (c = 0.39 p = 0.11). In contrast quantitative, molecular assessment of SSTR2 mRNA expression by PCR correlated very strongly with SUV mean (c = 0.85 p < 0.001) and equally well as SUV max itself did correlate with SUV mean (c = 0.90 p < 0.001). This demonstrates as proof-of-principle that target assessment by molecular in vitro and in vivo methods being quantitative by nature do perfectly fit for patient selection for imaging and potentially subsequent radionuclide treatment approaches.

However, tumor response to radionuclide treatments does not only dependent on total uptake, but also on tumor biological aspects such as intrinsic and neoplastic DNA repair capacity, proliferation status, hormone dependence, and tumor microenvironment. Precision oncology approaches have to take these complex interactions into account to improve completeness of therapy responses and thereby support long-term survival. As one example, the biology of the Prostate-Specific Membrane Antigen (PSMA) in prostate cancer might serve as being one of the most advanced radionuclide therapies. PSMA is a transmembrane glycoprotein, whose expression on prostate epithelium is of functional importance for cell migration and chromosome stability [25] and inversely regulated by androgens with increased activity found in tumor cells that become androgen-independent [26]. Superior efficacy of radioligand PSMA treatment (177Lu-PSMA-617) compared to standard of care in castration-resistant, metastatic prostate cancer previously treated with at least one androgen-receptor-pathway inhibitor and one or two taxane regimens and who had PSMA-positive gallium-68 (68Ga)-labeled PSMA-11 positron-emission tomographic-computed tomographic scans has been demonstrated [27]. Median overall survival reached 15.3 month for PSMA-targeted therapy versus 11.3 months for standard of care (Hazard ratio 0.62 p < 0.0001). Systematic review emphasizes clinical benefit for this radioligand therapy with 46% of patients achieving a reduction in PSA values >50% (and 75% had a decrease in PSA levels posttreatment) and an overall clinical benefit rate of 75.5% (37.2% of patients with PR and 38.3% SD) [28]. However, despite clear superiority over standard treatment, this study shows that singular radionuclide treatment still has limited efficacy in metastatic prostate cancer, as most patients progress and die of the disease. Molecular tissue analysis of repair genes such as BRCA1, BRCA2, ATM, CHEK2 may be one causal role for resistance or response to PSMA targeting with loss DNA-damage “recognition and signaling” genes resulting in resistance and loss of DNA-damage “repair” (such as BRCA2) being associated with increased radiosensitivity [29]. Interestingly, such “BRCAness” might be induced by PARP inhibition as has been shown in model systems [30]. Moreover, hormone receptors and signaling pathways (PTEN, AKT, PI3K, CDK1) contribute to development of resistance towards PARP inhibition [31], while PARP2 interacts with AR signaling, which in turn regulates PSMA expression. The multitude of functional interaction demonstrates that there is need of precise dissection of gene alteration, target quantitation and pathway pattern analytics in vitro to allow precise, multimodal approaches and adjusted therapy sequences, which combine radionuclide therapies with antihormonal, immune/vaccination therapies and simultaneous multitargeting by upcoming Antibody-Drug Conjugates (ADC). However, these therapeutic, multimodal approaches should in turn be monitored by molecular means again combining in vitro and in vivo approaches based on molecular assessment of tissue, urine, and blood diagnostics and pre- versus post-treatment imaging. Ultimately, these approaches shall not only be designed to govern direct tumor cell killing, but rather provoke systemic, longer lasting immune effects, that allow long-term survival. Most recently, we could show that long-term survival in metastatic NSCLC treated with first-line pembrolizumab monotherapy could be predicted after first cycle by quantitation of dynamic changes of immune cell mRNA signatures from peripheral blood pre- versus post-treatment [32]. Such approaches provide new early outcome indicators and may therefore be helpful to accelerate adopted precision oncology strategies and underline the importance of inducing immune responses in the advanced treatment settings. In summary, molecular research in the past decades pave the way for fundamentally new insights and treatment approaches with combined molecular in vitro and in vivo diagnostics emerging as the impartible basis of upcoming, multimodal therapy approaches in precision oncology.