Genetically engineered, conditionally replicating herpes simplex viruses type 1（HSV-1） are promising therapeutic agents for cancer. We have developed a triple-mutated, third-generation oncolytic HSV-1, G47Δ, that exhibits enhanced replication capability in a variety of cancer, efficient induction of specific antitumor immunity, and high safety features. G47Δ also kills cancer stem cells very efficiently.
The first-in-man, phase I-IIa study of G47Δ in recurrent glioblastoma patients was performed from 2009 to 2014. For each patient, G47Δ was administered stereotactically into the tumor, twice to the same coordinates within 14 days. This first trial proved the safety of G47Δ when injected into the human brain. The short-term efficacy due to replication and direct oncolytic activity of G47Δ was observed in almost all patients. The long-term efficacy seemed to have been caused by the induction of specific antitumor immunity rather than the oncolytic activity of G47Δ. WHO criteria were not suited for evaluating the efficacy of oncolytic virus therapy, because of pseudo-progression of tumors that always occurred immediately after G47Δ injection.
The phase II investigator-initiated clinical trial in patients with recurrent or residual glioblastoma started in 2015 to test the efficacy. In this phase II, each patient received stereotactic injections with G47Δ into the tumor, repeatedly, but into different coordinates, every 4 weeks, for the maximum 6 times. A set dose of 1 x 10E9 pfu/dose was used. In most patients, G47Δ was given in addition to the standard maintenance therapy using temozolomide. The interim analysis showed that the 1-year survival rate of 13 patients, the primary endpoint of this trial, was 92.3%, which was significantly higher than 15%, the set control value based on meta-analysis of historical data. The statistical significance was above the criteria of early termination, so this trial was terminated early and considered as the pivotal study for new drug application. Because G47Δ was designated as a ‘ Sakigake’ breakthrough therapy drug by the Japanese government, and further as an orphan drug for malignant glioma, a fast-track drug approval is expected.
G47Δ has proven safe even when injected repeatedly 6 times into the brain. The use of G47Δ may become a standard treatment that potentially leads to a cure of malignant glioma in the near future. Besides glioblastoma, we started a phase I study in prostate cancer patients and another study in olfactory neuroblastoma patients, both in 2013. The former was recently completed, which again proved the safety of G47Δ when injected into the prostate. A phase I study for malignant pleural mesothelioma has recently started, in which G47Δ is injected repeatedly into the pleural cavity. Advancing with glioblastoma trials was our strategy to obtain the earliest drug approval of G47Δ in Japan. Further, because G47Δ should be useful for a variety of solid cancer, we intend to expand the clinical application of G47Δ as swiftly and efficiently as possible.

Acral melanoma was the main subtype in Chinese patients（pts）, which was prone to intralymphatic diseases available for intra-lesional injection. The safety and efficacy of OrienX010, a herpes simplex virus type 1-derived oncolytic immunotherapy with expression of gene encoding human GM-CSF, was tested in phase Ⅰb trial of unresected stage ⅢC to Ⅳ（M1a/M1b） acral melanoma pts. The safety of OrienX010（8 × 10^7 pfu/ml） at ≤ 4 × 10^8 pfu q2w according to size of injectable lesions（1-10cm） was tested in phase 1b 08. If there was no DLT, the maximum dose would be escalated to ≤ 8 × 10^8 pfu q2w in phase 1b 09. Objective response rate（ORR） was evaluated q12w per iRECIST criteria. DLT was any grade（gr） ≥ 4 adverse event（AE） or gr ≥ 3 immune-related AE within first 6 wks. Between June 2014 and May 2017, 26 patients were enrolled in phase 08（n=12）, then phase 09（n=14）. Time to response was 4-12 weeks and median PFS was 20.00wks（95%CI 10.04-29.96）. The ORR was 23.08%（n=6）, CBR 57.69%（n=15） and DOR（6 months） 23.08%（n=6） in total. Median PFS was 12.00 wks（95%CI: 9.32-14.68 weeks） in group 08 and 24.00 wks（ 95% CI: 10.04-29.96 weeks） in group 09. Higher-dose showed relatively shorter time to response and longer PFS. Fever（73.08%）, injection site reaction（69.23%） and proteinuria （26.92%） were the most frequently AEs and no DLT were recorded. Randomized（2 : 1） phase 2 trial OrienX010 with the dose of 8×10^8 pfu vs. DTIC are now ongoing.
Liver metastases are relatively common in acal and mucosal melanoma, which confer a very poor prognosis and limited treatment options. After the clinical trial of intralymphatic disease, a phase Ic study of intralesional injection of OrienX010 into liver metastases in melanoma pts was conducted. Thirty stage IV M1c melanoma pts with injectable liver metastases patients were enrolled from May 2016 to September 2017. Ten were recruited as a safety cohort, and followed by another twenty pts as an expansion cohort. Up to 10 ml of 8 × 10^7 pfu/ml OrienX010 was injected intra-lesionally by ultrasound guidance every two weeks. The dose was distributed into 1-2 liver lesions according to tumor location and size. The primary endpoint was safety. Secondary endpoints included ORR, DCR, and PFS. The median follow-up time was 13 months（range, 2-26 months）. Pts received an average of 9.8 injections（range, 2-31）. Common AEs included pyrexia（100%）, nausea（30.0%）, leukopenia（30.0%）, chills（26.7%）, elevated blood transaminase（26.7%）, hyperbilirubinemia（10.0%）, and fatigue （10.0%）. All AEs were mild to moderate except one grade 3 pyrexia. Twenty-seven patients were evaluable by RECIST 1.1, ORR and DCR were 7.4% and 33.3% respectively, 22.2%（6/27） exhibited a durable response over 24 weeks. For target lesions in the liver, DCR reached 44.4%（12/27）. The one-year survival rate was 71.4%, and the median overall survival（OS） was not reached.
Oncolytic virotherapy with OrienX010 injections into intralymphatic disease were well tolerated and demonstrated similar efficacy like in Caucasians, and for M1c pts with liver metastasis, it also presents a safe and prospective treatment option. Clinical trials of combination therapy with anti PD-1Ab are now pending.

Our discovery of EML4-ALK oncogene in lung cancer brought highly effective ALK inhibitors to the patients carrying this fusion gene. Identification of other ALK fusions in various cancer subtypes further led to realize the importance of beyond-organ, gene-dependent cancer classification scheme. For ALK, such tumors oncogenic ALK may be collectively called “ALKoma”. On the other hand, many oncogenes have been discovered in lung cancer in a mutually exclusive manner, and it is now clear that lung cancer is never homogenous but amalgamation of many subtypes each carrying its specific oncogene. These lines of information have greatly contributed to bring the“ cancer genomic medicine” to the clinics. To assist cancer genomic medicine, we have developed a TOP panel for cancer multigene test that can sensitively and reliably detect gene fusions, in addition to single nucleotide variations, from formalin-fixed specimens. Further, to functionally annotate variants of unknown significance（VUS） in the cancer genome, we have created MANO method that screen hundreds of VUS in a high-throughput manner.
Japan is likely going to soon adapt gene-panel testing to optimize cancer treatments under national health insurance system. To discuss a necessary platform to perform such cancer genomic medicine in Japan, The Expert Meeting for Cancer Genomic Medicine Promotion Consortium was held in the Spring of 2017 in The Ministry of Health, Labour and Welfare. The Expert Meeting recommends a step-wise adaptation of genomic medicine, i.e. such medicine should be first conducted only in designated hospitals, and the number of these hospitals shall be increased gradually. Another important proposal from the Expert Meeting is to set a central datacenter to aggregate genomic as well as clinical information of gene-panel tests. The Ministry of Health, Labour and Welfare accordingly designated, in the Spring of 2018, eleven Core Hospitals for Cancer Genomic Medicine and 156 of Cooperative Hospitals for Cancer Genomic Medicine. The Ministry also established, in June 2018, The Center for Cancer Genomics and Advanced Therapeutics （C-CAT） to store and utilize genomic/clinical information. The genomic medicine platform within Japan, consisting of these hospitals and C-CAT will be further discussed in the symposium.

Lung cancer is the leading cause of cancer-related death. It is often diagonized at advanced disease and chemotherapy used to be its standard therapy associated with marginal improvement of overall survival. Now lung cancer could be further divided into several subsets according to oncogenic drivers, such as EGFR mutation, ALK rearrangements, ROS1 rearrangements, KRAS mutation, etc. Tumor molecular profiling is a fundamental component of precision oncology, enabling the identifi cation of genomic alterations in genes and pathways that can be targeted therapeutically. Several targeted therapies have come into our clinical practice and significantly improve outcomes of the patients.
EGFR TKIs become standard 1st line therapy for the patients with positive EGFR mutation. 1st, 2nd and 3rd generations of EGFR TKI prove effective in 1st line treatment of EGFR mutant NSCLC with the longest PFS seen with osimertinib. Better sequence of these TKIs is not clear. Combination of EGFR TKI and chemotherapy or bevacizumab was found to prolong PFS compared with TKI alone. But we do not know which patients should be treated with EGFR TKI + chemotherapy or bevacizumab. EGFR T790M accounts for 60% of acquired resistance to 1st or 2nd generation of EGFR TKIs. Osimertinib led to longer PFS than chemotherapy in those with positive T790M mutation previously treated with EGFR TKI. NGS reveals that over 50% of EGFR mutant NSCLC patients have companion genetic mutations. There companion mutations affect efficacy of EGFR TKI. We need to establish therapy for these patients. The patients with uncommon EGFR mutation are usually refractory to EGFR TKIs. Several novel compounds are under clinical studies. The molecules responsible to acquired resistance to TKIs could be detected about 3 months before clinical evidence of disease progression. We do not know earlier interference at the time of detecting these molecules is better than at the time of clinical disease progression.
ALK TKI becomes standard 1st therapy for the patients with positive ALK. 2nd generation of ALK TKIs do better than crizotinib in terms of PFS and brain metastasis control. PFS with 1st line alectinib is about 3 years, longer than crizotinib. Loratinib was effective after failure of crizotinib or two ALK TKIs. Molecular mechanisms associated with 2nd or 3rd generation ALK TKI are not clear.
In near future, we establish targeted therapy for the patients with rare oncogenic drivers, such as ROS1, RET, NTRK, c-MET exon 14 skipping mutations.
In summary, majority of non-squamous NSCLC patients benefit from targeted therapies and their outcome is improved. Some novel compounds with better efficacy and less toxicity will be available in the markets. Mechanisms associated with novel targeted agents will be understood and then targeted therapy against these mechanism will be developed.

Quantum physics and related engineering have changed the world in many fi elds in 20th century. Cancer treatment is one of them. Accelerators of electrons, protons, and carbons are now used in radiotherapy curing many cancers from the head to toe often without requirement of surgical operations and toxic chemotherapy. At present, many cancer patients are treated by the radiotherapy in the world to soften the symptoms in most occasions and completely cure the disease in many occasions. Majority of the cancer patients treated by the radiotherapy have enjoyed their usual quality of life and not required to leave their jobs because of the minimal invasiveness during and after the treatment.
However, at present, there are a portion of cancer patients with large and resistant cancers who still require unnecessary toxic irradiation to the surrounding normal tissue. The guiding principle in radiotherapy, therefore, is the optimization in the trade-off between dose to the target（cancer） and the critical organs at risk（OAR） in the surrounding normal tissue. Precise localization using image-guidance and dose calculation in one milli-meter accuracy with a consideration for organ motion have improved the trade-off in the last 30 years.
Recent evidence about the success in the combination of immune-check point blockage with radiotherapy strongly suggest that it is so important to reduce the immunosuppressive effects caused by conventional cancer ‘treatment’. In near future, we need to change our guiding principle of radiotherapy to include the immunosuppression as one of the critical adverse effect. We may improve our local control rate of radioresistant cancers by reducing the dose to the OAR regarding immunosuppression. The problem is that it is not still clear where is the OAR regarding immunosuppression in each patient. Our animal study has suggested that it is better to preserve regional lymph-nodes when we irradiate the primary tumors（Takeshima, Chamoto, et al. Cancer Res. 2010）. Recent real-time-image gated radiotherapy can irradiate a moving lymph-nodes with 1-mm accuracy if the position of the critical part of the lymph-nodes can be visualized using fiducial markers during the irradiation（Shimizu, et al. Plos One 2014）. There is strong motivation in the radiation oncology society to find the way to visualize the OAR for immunosuppression before and during the radiotherapy. International collaboration in biomedical science and engineering is required to find the best way to preserve the OAR for immunosuppression in radiotherapy.

To date, several kinds of immunomodulating monoclonal antibodies （mAbs） have been applied in clinical trials, such as anti-cytotoxic T-lymphocyte antigen-4 （anti-CTLA-4） mAb and anti-programmed death-1（anti-PD-1） mAb. With the recent success of cancer immunotherapy, especially the checkpoint inhibitors, the renewed interest in immunotherapy as a treatment modality has gained extensive attention. The irradiated tumor cell death can enhance antitumor immunity by inducing antigen expression on tumor cells and activating lymphocytes. Radiotherapy（RT） combined with immunotherapy has revealed promising outcomes in various animal models. However, this new paradigm is often considered as a medical spectacle without a unifying model, and its mechanisms have yet to be elucidated. The following challenges need to be elucidated: exploring the potential mechanisms; identifying the most beneficial dose, fraction and target site for RT; finding an appropriate time window to combine these two treatments; and discussing the toxicity and suitable treatment evaluating criteria. Besides, establishment of predictive biomarkers for checkpoint blockades is of the most importance to maximize the therapeutic benefits. PD-L1 expression is a logical biomarker for the prediction of response to the anti-PD-（L）1 immunotherapy. However, the predictive values of PD-L1 expressions for immunotherapy are currently debating and challenging. Multiplex detecting methods and combined biomarkers may provide new strategies. For example, tumor mutation and neoantigens burden, some oncogene mutations, like EGFR, ALK, KRAS and STK11. In addition, with development of new probes and tracers, immuno-PET provides a new, non-invasive and quantitative strategy to monitor treatment response. As current evidence of those potential
predictors, a consensus and standardization is needed to establish to widely applied in future studies. Multiplex detecting methods and combined biomarkers may provide new strategies.

The magnifying colonoscopy has been used to diagnose colorectal lesions for more than 10 years. Pit pattern classification with chromo-magnifying endoscopy and narrow-band image（NBI）system are useful for differential diagnosis of a colorectal lesion, and for predicting the depth of cancer invasion. For more precise diagnosis, we proposed “endocytoscopic classification” and “EC-V pattern” using endocytoscopy（EC） with a magnification power of about × 520. This super-magnifying endoscope enables the direct assessment of the cell and the nucleus.
However, to diagnosis the colorectal lesion precisely based on the endoscopic findings needs training and sometimes influenced by the diagnostic ability of each endoscopists. To solve this issue, we developed the computer-aided diagnosis（CAD）system for EC（EndoBRAIN®）.
The EndoBRAIN system diagnose the colorectal lesions based on the obtained about 300 featured from nuclei, vessels, and texture analysis of EC images. EndoBRAIN is directly connected to an endoscopic unit and obtained EC images are transferred to the system automatically. EndoBRAIN present the predicted pathological diagnosis with probability using diagnosing algorithm installed to the system and it takes only 0.4 seconds. Endoscopists can use the EndoBRAIN diagnosis as a reference during observation of colorectal lesions. We have reported the usefulness of the EndoBRAIN for colorectal lesions. The EndoBRAIN system launched in Japan since March 2019. A lesion detection support system based on deep learning is under construction but it will hit the market place in near future.
The combination of super-magnifying endoscopy and AI diagnosis will provide the next stage of endoscopic diagnosis for all endoscopists.

China has a high incidence and death rate of esophageal cancer in the world. There are about 300,000 people died of esophageal cancer around the world every year, and half of them happened in China. There is a special prevalence mode in China that most of the high incidence areas are surround Taihang mountains area. Squamous cell cancer accounts for about 90％ of all esophageal cancer cases in China. Endoscopic therapies usually used for superficial esophageal cancer and Intraepithelial neoplasia（ESCN）include endoscopic submucosal dissection（ESD）, endoscopic mucosal resection（EMR）, multi-band mucosectomy（MBM）, radiofrequency ablation（RFA）and cryo ablation. Piecemeal resection technology such as EMR and MBM has been widely used in China for ESCN, but disadvantages could not be ignored such as higher occurrence rate of the postoperative stricture of esophagus if the length of the lesion is longer than 4cm or 50％ of the cross-section. Moreover, this technology should be operated by skilled endologists and it brings difficulty for the biopsy diagnose because of more pieces.
Nonetheless, MBM is still eff ective for the complete removal lesions with short procedure time and few complications. Apart from resection technology, endoscopic radiofrequency ablation（RFA）is another effective treatment method for moderate and high-grade intraepithelial squamous neoplasia, cryo ablation is also a selective option for intraepithelial squamous neoplasia of esophagus.